preparation and characterization of biopolymer compounds
TRANSCRIPT
Preparation and characterization of biopolymer compounds
containing poly-3-hydroxyalkanoates and polylactic acid
by
Manoj Nerkar
A thesis submitted to the Department of Chemical Engineering
In conformity with the requirements for the degree of
Doctor of Philosophy
Queen’s University, Kingston, Ontario, Canada
(September, 2014)
Copyright © Manoj Nerkar, 2014
i
Abstract
This thesis is focused on developing cost effective and environmentally friendly techniques to
improve the properties and processability of biopolyesters through compounding and reactive
modification. Specifically, elastomeric medium-chain-length poly(3-hydroxyalkanoates) (MCL
PHA) have been evaluated as potential impact modifiers for poly(lactic acid) (PLA) and poly-3-
hydroxybutyrate (PHB), using conventional melt compounding. The Mark-Houwink constants,
absolute molecular weight distributions and the absolute molecular weight (MW) averages of
MCL PHAs with predominantly 3-hydroxyoctanoate (PHO), 3-hydroxynonanoate (PHN) or 3-
hydroxydodecanoate (PHDD) content were determined and ranged between 18,200 for PHN to
172,000 Da for PHO. Detailed thermal and rheological characterization revealed that PHO had
the highest viscosity, and was thus the best candidate as impact modifier for PHB and PLA. Melt
compounded PHB/PHO and PLA/PHO blends showed improved tensile strain at break and
unnotched impact strength upon addition of up to 30 wt.% PHO in PHB and 15 wt.% PHO in
PLA. This was counteracted by decreased Young’s modulus due to lower blend crystallinity. The
droplet-matrix morphology coarsened as PHO content increased beyond 5 wt.%, due to PHO
coalescence attributed to viscosity mismatch between blend components. PHO was reacted
using lauroyl peroxide to increase its viscosity through partial cross-linking, thus improving the
morphology but the mechanical properties showed only moderate improvements, presumably
due to high PHO gel content which compromised its elastomeric nature. Reactive compounding
by radical mediated solvent-free grafting of triallyl trimesate (TAM) coagent was employed to
improve blend properties. Reactively modified PLA had higher molar mass, melt viscosity and
enhanced strain hardening. Additionally it showed a distinct crystallization peak upon cooling
ii
with disappearance of the cold crystallization peak, indicative of a nucleation effect. PLA
modified using a multi-functional epoxide oligomeric chain extender yielded similar
improvements in rheological properties, but no considerable change in crystallization. This
coagent modification approach also increased the viscosity of PHO, and improved both
extrudate appearance and handling. Coagent modified PLA/PHO blends demonstrated
significant improvement in crystallization and rheological properties, similar to those seen in
the coagent modified PLA alone, while the mechanical properties remained unaffected.
iii
Co-Authorship
This thesis contains seven chapters that present results that have been published in the form of
original journal articles as well as material that is in preparation for submission. The complete
citations for the published papers and chapter in which they appear are provided bellow:
Chapter 3: Nerkar M, Ramsay J, Ramsay B, Kontopoulou M, Hutchinson R.
Determination of Mark-Houwink parameters and absolute molecular weight of medium-
chain-length poly(3-hydroxyalkanoates). Journal of Polymers and the Environment 2013
(21): 24-29
Chapter 4: Nerkar M, Ramsay J, Ramsay B, Kontopoulou M. Melt compounded
blends of short and medium-chain-length poly-3-hydroxyalkanoates. Journal of
Polymers and the Environment 2014 (22): 236-243
Chapter 5 : Nerkar M, Ramsay J, Ramsay B, Kontopoulou M. Dramatic Improvements
in Strain Hardening and Crystallization kinetics of PLA by simple reactive modification in
the melt state. Macromolecular Materials and Engineering (Accepted –May 2014)
Chapter 6 : Nerkar M, Ramsay J, Ramsay B, Kontopoulou M. Improvements in the
extensional rheology, thermal properties and morphology of poly(lactic acid)/ poly-3-
hydroxyoctanoate blends through reactive modification (To be submitted)
All the papers and manuscripts were co-authored and reviewed prior to publication by
Professors Marianna Kontopoulou, Juliana A. Ramsay and Bruce A. Ramsay. The first paper
(chapter 3) was co-authored by Professor Robin Hutchinson who directed the experiments to
determine true molecular weight of medium-chain-length polyhydroxyalkanoate. Dr.
Alexandros Vasileiou helped in the synthesis and characterization of epoxidized
iv
polyhydroxyoctanoate presented in Appendix A. Hang Li performed fermentations that made
medium-chain-length polyhydroxyalkanoates. All the rest of the experimental work and
manuscript preparation were performed by the author of this thesis.
v
Acknowledgements
It would not have been possible to complete my doctoral thesis without the support and help
from my supervisors, staff members, lab mates, family and friends.
I consider myself lucky to have a supervisor like Professor Marianna Kontopoulou. I appreciate
her for being a great support to me and to my family in a foreign land. Apart from her endless
technical support and research guidance, she was always motivating and encouraging. She has
been a wonderful person to work with. It’s my great privilege to be co-supervised by Professor
Juliana A Ramsay. I would like to thank her for her technical guidance and for taking time to
critically reviewing all my work. I was privileged to have the opportunity to work closely with
Dr. Bruce A Ramsay. He was instrumental in providing his expert opinion about biopolymers. I
would like to thank him, along with his team members, Dr. Zhiyong Sun, Hang Li and Eric Potter
for providing the polymers required for my research.
Professor Robin Hutchinson is greatly acknowledged for his support in molecular weight
characterization of polymers. Timely support from Dr. Kalam Mir, Steven Hodgson and Kelly
Sedore is highly appreciated. Charlie Cooney from the materials department provided judicious
help in conducting SEM analysis of samples.
I would like to thank my lab mates Osayuki Osazuwa, Ying Zhang, Andrew Powell, Praphulla and
Dr. Alexandros Vasileiou for their friendship, technical discussions and help.
Funding support from the Natural Science and Engineering Research Council (NSERC), Queen's
University Graduate Award and Xerox Research Center of Canada is gratefully acknowledged.
I take this opportunity to thank my parents Mrs. Rajani Gokul Nerkar, Mr. Gokul
Madhavrao Nerkar, and my in-laws, Mrs. Shailaja Arvind Sonar and the late Mr. Arvind
vi
Lilachand Sonar and all family members including brothers, brother-in-laws and sister-in-laws
and friends from around the world who provided all kinds of support throughout my life.
Poonam, my lovely wife, is the person who inspired me to get doctorate. I cannot thank her
enough for all of her sacrifices and for being with me all the time. She took care of our kids and
their illness single handedly when I was busy studying for exams, attending conferences or
writing manuscripts.
My acknowledgement cannot be completed without mentioning my adorable son Aayush and
my beautiful daughter Anishka. I thank them for their unconditional love.
vii
Contents
Abstract ...................................................................................................................................................... i
Co-Authorship .......................................................................................................................................... iii
Acknowledgement .................................................................................................................................... v
List of Figures ............................................................................................................................................ x
List of Tables ...........................................................................................................................................xiii
List of Schemes........................................................................................................................................ xiv
Nomenclature ......................................................................................................................................... xv
Abbreviations .......................................................................................................................................... xvi
Chapter 1 Introduction .............................................................................................................................. 1
1.1 Thesis objectives ........................................................................................................................... 2
1.1.1 Characterization of MCL PHA ................................................................................................ 3
1.1.2 Blends of MCL PHA with brittle biopolymers ........................................................................ 3
1.1.3 Enhancement of melt viscosity of MCL PHA for blending with PHB and PLA ....................... 3
1.1.4 Improved melt strength and crystallization of PLA and its blends with MCL PHA by
reactive modification ............................................................................................................................ 4
1.2 Thesis organization ....................................................................................................................... 4
Chapter 2 Literature Review ...................................................................................................................... 6
2.1 Polyhydroxyalkanoates ................................................................................................................. 6
Medium-chain-length (MCL) PHA ......................................................................................................... 7
2.1.1 Impact modification of PHB .................................................................................................. 8
2.1.2 Plasticization ......................................................................................................................... 9
2.1.3 Nucleation ........................................................................................................................... 10
2.1.4 Chain extension ................................................................................................................... 12
2.2 Polylactic acid (PLA) .................................................................................................................... 12
2.2.1 Impact modification ............................................................................................................ 14
2.2.2 Plasticization ....................................................................................................................... 14
2.2.3 Nucleation ........................................................................................................................... 15
2.2.4 Conditioning ........................................................................................................................ 15
2.2.5 Chain extension ................................................................................................................... 16
2.2.6 Epoxy based chain extension .............................................................................................. 16
viii
2.2.7 Peroxide-mediated cross-linking......................................................................................... 18
2.2.8 Multifunctional coagents .................................................................................................... 19
Chapter 3 Determination of Mark-Houwink parameters and absolute molecular weight of medium-
chain-length poly(3- hydroxyalkanoates)* ................................................................................................. 21
3.1 Introduction ................................................................................................................................ 21
3.2 Experimental ............................................................................................................................... 23
3.2.1 Materials ............................................................................................................................. 23
3.2.2 Methods .............................................................................................................................. 24
3.3 Results and Discussion ................................................................................................................ 26
3.3.1 Molecular weight determination ........................................................................................ 26
3.3.2 Thermal and rheological characterization .......................................................................... 32
3.4 Conclusion ................................................................................................................................... 34
Chapter 4 Melt compounded blends of short and medium-chain-length poly-3-hydroxyalkanoates* .. 35
4.1 Introduction ................................................................................................................................ 35
4.2 Experimental ............................................................................................................................... 37
4.2.1 Materials ............................................................................................................................. 37
4.2.2 Compounding ...................................................................................................................... 37
4.2.3 PHO cross-linking ................................................................................................................ 37
4.2.4 Blend Characterization ........................................................................................................ 38
4.3 Results and discussion ................................................................................................................ 40
4.3.1 Thermal and rheological properties .................................................................................... 40
4.3.2 Morphology ......................................................................................................................... 44
4.3.3 Mechanical properties ........................................................................................................ 46
4.3.4 Cross-linking of PHO ............................................................................................................ 47
4.4 Conclusions ................................................................................................................................. 52
Chapter 5 Dramatic improvements in strain hardening and crystallization kinetics of PLA by simple
reactive modification in the melt state*..................................................................................................... 54
5.1 Introduction ................................................................................................................................ 54
5.2 Experimental ............................................................................................................................... 56
5.3 Results and Discussion ................................................................................................................ 59
5.3.1 Rheological characterization ............................................................................................... 59
5.3.2 Thermal properties ............................................................................................................. 63
ix
5.4 Conclusions ................................................................................................................................. 66
Chapter 6 Improvements in the extensional rheology, thermal properties and morphology of
poly(lactic acid)/ poly-3-hydroxyoctanoate blends through reactive modification ................................... 68
6.1 Introduction ................................................................................................................................ 68
6.2 Experimental ............................................................................................................................... 70
6.2.1 Materials ............................................................................................................................. 70
6.2.2 Compounding ...................................................................................................................... 70
6.2.3 Characterization .................................................................................................................. 71
6.3 Results and Discussion ................................................................................................................ 74
6.3.1 Blends of PLA with PHO ...................................................................................................... 74
6.3.2 Reactive modification of PHO ............................................................................................. 77
6.3.3 Reactive modification of PLA .............................................................................................. 81
6.3.4 Reactive compounding of PLA with PHO ............................................................................ 85
6.3.5 Thermal and rheological properties .................................................................................... 86
6.3.6 Blend morphology ............................................................................................................... 86
6.3.7 Mechanical properties ........................................................................................................ 89
6.4 Conclusions ................................................................................................................................. 91
Chapter 7 Thesis overview ....................................................................................................................... 92
7.1 Thesis overview ........................................................................................................................... 92
7.2 Conclusions ................................................................................................................................. 93
7.3 Significant contributions ............................................................................................................. 96
7.4 Recommendation for future work .............................................................................................. 97
References .................................................................................................................................................. 99
Appendix A - Improved viscosity ratio and compatibility of poly (lactic acid) and polyhydroxyoctanoate
blends ........................................................................................................................................................ 114
x
List of Figures
Figure 3.1 a) Experimental Mark-Houwink data for PHO (three replicates), b) best fit Mark-
Houwink relationships for all copolymer samples. ....................................................................... 28
Figure 3.2 Molecular weight distributions for PHO (three replicates) as measured by a)
Waters/Wyatt SEC, light-scattering detector; b) Waters/Wyatt SEC, universal calibration; c)
Viscotek SEC, triple detection; and d) Viscotek –SEC, universal calibration. ............................... 30
Figure 3.3 Molecular weight distributions of four MCL PHA polymers (Viscotek SEC, universal
calibration). ................................................................................................................................... 32
Figure 3.4 Viscosity as a function of molecular weight ................................................................ 33
Figure 4.1 TGA curves for PHO, PHB and the 85/15 PHB/PHO blend at 190 °C ........................... 41
Figure 4.2 a) DSC endotherm (2nd heating cycle) and b) DSC exotherm of PHB and PHB/PHO
blends ............................................................................................................................................ 43
Figure 4.3 Rheological properties of PHB and PHO and effect of peroxide cross-linking on a)
complex viscosity, b) storage modulus and c) loss tangent, tan δ, measured at 190 °C ............. 45
Figure 4.4 Scanning electron microscopy blends containing a) 5 wt.%, b) 10 wt.%, c) 15 wt.%, d)
20 wt.% and e) 30 wt.% of PHO at 2000x magnification .............................................................. 46
Figure 4.5 a) Un-notched impact strength and tensile strain b) tensile stress and Young’s
modulus of pristine PHB and its blends ........................................................................................ 48
Figure 4.6 Cure curves of PHO with 0.1 wt.% lauroyl peroxide as a function of temperature at a
frequency of 1 Hz. ......................................................................................................................... 49
xi
Figure 4.7 SEM of blends of PHB/cross-linked PHO 70/30 blends a) uncross-linked (viscosity
ratio, λ=0.03) b) cross-linked with 0.06 wt.% of lauroyl peroxide (λ=0.15) c) cross-linked with 0.2
wt.% of lauroyl peroxide (λ=0.36) d) cross-linked with 0.5 wt.% of lauroyl peroxide (λ=3.73). .. 51
Figure 5.1 a) Complex viscosity as a function of frequency and b) phase degree as a function of
complex modulus at 180 °C. ......................................................................................................... 60
Figure 5.2 Tensile stress growth coefficient (ηE+) of TAM and GMA modified PLA as a function of
strain rate and time at Hencky strain rates of 0.1, 1 and 10 s-1 at 180 °C. Curves are shifted by an
arbitrary factor for the sake of clarity. Solid lines represent the LVE envelop (3E+) for each
sample. .......................................................................................................................................... 62
Figure 5.3 DSC a) 2nd heating scan at rate of 5 °C/min b) cooling scan at the rate of 5 °C/min . 63
Figure 5.4 Relative degree of crystallinity as a function of time a) isothermal crystallization
experiments; (-) PLA/TAM at 135 °C, ()PLA/TAM at 140 °C, ()PLA/TAM at 150 °C, (◆)
PLA/GMA at 135 °C and (b) non-isothermal crystallization experiments; ()PLA/TAM at 2.5
°C/min, (◆)PLA/TAM at 5 °C/min, (o)PLA/TAM at 20 °C/min, ()PLA/GMA at 2.5 °C/min,
()PLA/GMA at 5 °C/min .............................................................................................................. 65
Figure 6.1 Scanning electron microscope images of PLA blend containing a) 5 wt.%, b) 10 wt.%,
c) 15 wt.% and d) 20 wt.% of PHO. ............................................................................................... 75
Figure 6.2 Effect of TAM content on the rheological properties of PHO with DCP content
remaining constant a) Complex viscosity b) storage modulus and c) tan δ ................................. 78
Figure 6.3 a) unmodified PHO after extrusion b) PHO/0.3/1 after extrusion .............................. 79
Figure 6.4 Effect of DCP amount on a) Complex viscosity b) storage modulus and c) tan δ of
coagent modified PHO (PHO 0.3/1 and PHO 0.5/1 yielded 23 and 42 % gel respectively) ....... 80
xii
Figure 6.5 Effect of coagent modification on the complex viscosity of PLA and PHO ................. 81
Figure 6.6 Tensile stress growth coefficient (ηE+) of PLA/0.3/1 and (PLA/PHO)/0.3/1 as a
function of strain rate and time at Hencky strain rates of 0.1, 1 and 10 s-1 at 180 °C. Curves are
shifted by an arbitrary factor for the sake of clarity. Dotted lines represent the LVE envelop for
each sample. ................................................................................................................................. 82
Figure 6.7 DSC (a) cooling exotherm (b) heating endotherm of coagent-modified PLA and
PLA/PHO blends ............................................................................................................................ 84
Figure 6.8 Hot stage microscopy of a) PLA, b) PLA/0.3/1 at 135 °C ............................................. 85
Figure 6.9 Effect of DCP and TAM on a) complex viscosity b) storage modulus and c) tan δ of
PLA/PHO blends ............................................................................................................................ 87
Figure 6.10 Scanning electron microscopy of PLA/PHO (90/10) blend a) unmodified b)
(PLA/PHO)/0.5 c) (PLA/PHO)/1 ..................................................................................................... 88
Figure 6.11 Effect of coagent modification on morphology of PLA/PHO blends a) 95/05 b) 90/10
c) 80/20 (wt./wt.%) (samples reacted with coagent were not etched); Top raw without coagent;
bottom raw with coagent ............................................................................................................. 89
xiii
List of Tables
Table 2.1 Properties of PLA [37] ................................................................................................... 13
Table 3.1 Mark Houwink calibration constants from best fit to triple detector analysis of MCL
PHA samples in THF ([Ƞ]=K(MW)a). .............................................................................................. 29
Table 3.2 Number-average (Mn), weight-average (Mw), and dispersity (PDI) values determined
from SEC analysis of MCL PHA samples. ....................................................................................... 31
Table 3.3 Thermal properties of MCL PHAs .................................................................................. 33
Table 4.1 Crystallization temperature (TC), first and second melting peaks (TM1 and TM2
respectively), % crystallinity and degradation onset temperature for PHB and PHB/PHO blends.
....................................................................................................................................................... 42
Table 4.2 Comparison of mechanical properties of PHB/PHO 70/30 blends containing uncross-
linked and cross-linked PHO ......................................................................................................... 52
Table 5.1 Material characterization .............................................................................................. 61
Table 5.2 Isothermal Avrami constants and crystallization half time for PLA/GMA and PLA/TAM
at various temperatures ............................................................................................................... 66
Table 6.1 Mechanical properties of PLA and PLA/PHO blends ..................................................... 76
Table 6.2 Thermal properties of neat, DCP and coagent modified PHO, PLA and PHO/PLA blend
....................................................................................................................................................... 83
Table 6.3 Mechanical properties and heat deflection temperature of neat and coagent modified
PLA, and PLA/PHO blends ............................................................................................................. 90
xiv
List of Schemes
Scheme 2.1 Chemical structure of polyhydroxyalkanoate [8] ........................................................ 6
Scheme 2.2 Reaction of diisocyanate with hydroxyl and carboxyl functional groups, adapted
from Lee et al. 2009 [4] ................................................................................................................. 12
Scheme 2.3 Chemical structure of polylactic acid [36] ................................................................. 13
Scheme 2.4 Reaction between PLLA and glycidol adapted from Deenadayalan et al. 2009 [46] 16
Scheme 2.5 General structure of Joncryl® styrene – acrylic multi-functional oligomeric chain
extender, where R1 – R5 are H, CH3, a higher alkyl group, or combinations of them, R6 is an
alkyl group, and X, Y and Z are each between 1 and 20, adapted from Villalobos at al. 2004 [48]
....................................................................................................................................................... 17
Scheme 2.6 Mechanism of PLA reaction with GMA, adapted from Al-Itry et al. 2012 [51] ......... 18
Scheme 2.7 Chemical structure of coagents, adapted from Parent et al. 2008 [56] ................... 19
xv
Nomenclature
a Mark-Houwink exponent G* Complex modulus (Pa) G' Elastic or Storage modulus (Pa) G'' Viscous or loss modulus (Pa) K Mark-Houwink constant (dL/g) T Temperature (°C) t Time (s) TC Temperature of crystallization (°C) TM Melting point (°C) Xc Degree of crystallinity (%) Greek Symbols ΔHf Heat of fusion (J/g) [η] Intrinsic viscosity (dL/g) 3η+ Linear viscoelastic envelope in uniaxial extension (Pa.s) η* Complex Viscosity (Pa.s) η0 Zero shear viscosity (Pa.s) ω Angular frequency (rad/s) λ viscosity ratio ηd viscosity of the dispersed phase ηm viscosity of the matrix
xvi
Abbreviations
3HV 3-hydroxyvalerate ASTM American society for testing materials CHS cross head speed DBP Di n-butyl phthalate DCP Dicumyl peroxide dn/dc Refractive index DRI Differential refractive index DSC Differential scanning calorimetry GMA Glycidyl methacrylate GMS Glycerol monostearate GPC Gel permeation chromatography GTA Glycerol triacetate GTB Glycerol tributyrate HDI Hexamethylene diisocyanate HDT Heat distortion temperature IV Intrinsic viscosity LALS Low angle light scattering LLDPE Linear low density polyethylene LS Light scattering LVE Linear viscoelastic MCL Medium-chain-length mCPBA m-Chloroperbenzoic acid MK Mark-Houwink MMT Montmorillonite Mn Number-average MSDS Material safety data sheets Mw Weight-average MW Molecular weight MWD Molecular weight distribution NCC Nano-crystalline cellulose NSERC Natural sciences and engineering council PBS Polybutylene succinate PCL Polycaprolactone PDI Poly dispersity index PEG Polyethylene glycol PEGCA Poly (polyethylene glycol-co-citric acid) PEGCA Poly (polyethylene glycol-co-citric acid) PET polyethylene terephthalate PETA Pentaerythritol triacrylate PHA Poly-3-hydroxyalkanoate PHB poly(3-hydroxybutyrate
xvii
PHB-HV poly(3-hydroxybutyrate-co-3-hydroxyvalerate PHDD Poly-3-hydroxydodecanoate PHN Poly-3-hydroxynonanoate PHO Poly-3-hydroxyoctanoate PLA Poly (lactic) acid PLLA Poly(L-Lactic Acid) POM Poly(methyleneoxide PP Polypropylene PS Polystyrene PVA Poly vinyl alcohol RALS Right angle light scattering SEC Size exclusion chromatography SEM Scanning electron microscopy SNCC Silylated cellulose nanocrystals TAM Triallyl trimesate TAP Triallyl phosphate TC Crystallization temperature TD Triple detector Tg Glass transition temperature TGA Thermogravimetric analysis THF Tetrahydrofuran TM Melt temperature TMPTA Trimethylol propanetrimethacrylate TPEE Thermoplastic polyester elastomer UC Universal calibration
1
Chapter 1 Introduction
The rapid progress in polymer science and technology in the latter half of the twentieth century
has led to the use of synthetic polymers in almost every field, including household items,
industrial applications, electrical appliances, electronic devices, automotive, construction, and
the medical field. Polymer based compounds are found in every aspect of everyday life. The fast
growth of the polymer industry is attributed to the unique properties of polymers including
light-weight, corrosion resistance, design flexibility, and ease of manufacturing. Durability is one
of the advantages of polymers, but it is also their shortcoming. Petroleum based polymers are
generally not biodegradable. Their life span is longer than several hundred years. After the end
of their use, the polymers end up either in landfills or littering the environment. Landfilling is
becoming increasingly prohibitive, due to increasing costs, scarcity of land and other health and
environmental considerations, such as ground water contamination.
Furthermore, conventional plastics are made from petroleum products or natural gases, which
are non-renewable resources. Serious concerns about greenhouse gas emissions [1] and high oil
prices, which accompany the use of conventional polymers, are key driving forces to reduce
their usage. Therefore there is a need for new materials from renewable resources to replace
conventional polymers [2,3]. The quest for alternative materials has put biopolymers at the
forefront [4]. Biopolymers are promising candidates to replace conventional polymers, because
of their biodegradable nature and they can be made from renewable resources as raw material.
They can be categorized as a) bioresourced, b) biodegradable and c) bioresourced and
biodegradable.
2
Polylactic acid (PLA) is one of the major bioplastics available on a large commercial scale. It is an
aliphatic polyester made from α hydroxy acids. It is biodegradable and biocompatible and it is
used in many biomedical applications, such as sutures, stents, dialysis media, and drug delivery
devices. PLA is also used in commodity applications ranging from clothing, packaging, bottles,
and office stationary to food containers [5].
Polyhydroxyalkanoate (PHA) is another polymer that has drawn considerable attention. It is
produced by bacteria in a fermentation process, and is a water stable, biodegradable,
biocompatible polymer. Past efforts to commercialize PHA have not been very successful,
because they lack the engineering properties and processability needed to compete with
conventional polymeric materials [6].
1.1 Thesis objectives
Upgrading the properties of biopolymers is an active area of research in academia and industry,
with scientists and technologists striving to match performance of biopolymers with petroleum
based polymers to find new applications. The objective of this thesis is to develop commercially
viable biopolymer formulations containing PHA and PLA. To achieve this goal, medium-chain-
length (MCL) PHA has been identified as a potential impact modifier. Following the detailed
characterization and selection of suitable MCL PHA grades, challenges like the viscosity
mismatch between the polymers and slow crystallization rates will be addressed with the
ultimate goal of developing polymeric material with acceptable engineering properties and
good processability in commercial polymer processing operations.
The approach followed in this thesis is outlined below.
3
1.1.1 Characterization of MCL PHA
Even though the synthesis of MCL PHAs has been reviewed extensively, their physical
properties have not been yet fully characterized. Full characterization of the MCL PHAs used in
this study is necessary to further understand their physical properties and processability and to
choose suitable MCL PHA grades for formulations of biopolymers containing these materials.
The first objective of this work is to fully characterize the molecular weight, melt and solid state
properties of various MCL PHA grades, with different chain structures. This will aid with the
choice of materials that will be used in subsequent steps.
1.1.2 Blends of MCL PHA with brittle biopolymers
Biopolymers such as PLA and poly-3-hydroxybutyrate (PHB) are brittle materials and thus
cannot be used in certain applications like packaging, where flexibility of the material is
essential. The second objective is to assess the ability of MCL PHA to impart flexibility into PHB
and PLA. A suitable MCL PHA candidate will be chosen based on properties like melt viscosity,
melting temperature, thermal stability and molecular weight, as described in the first objective.
Various amounts of MCL PHA will be blended with PHB and PLA and the morphology and
mechanical properties of the blends will be evaluated.
1.1.3 Enhancement of melt viscosity of MCL PHA for blending with PHB and PLA
It is expected that the inherently low melt viscosity of MLC PHAs will pose a problem when
trying to blend them with other polymers. The significant difference in the melt viscosity of the
blend components tends to give phase separation, and coarse morphology, resulting in poor
mechanical properties. Therefore, the third objective will be to increase the melt viscosity of
MCL PHAs, using chain extension and cross-linking techniques.
4
1.1.4 Improved melt strength and crystallization of PLA and its blends with MCL
PHA by reactive modification
One of the main drawbacks that have hindered widespread implementation of biopolyesters,
including PLA and PHAs, is their low crystallization rates, resulting in poor mechanical
properties and processability. This makes them practically impossible to melt process in a cost-
effective way using conventional techniques like injection molding, compression molding and
extrusion. Poor melt strength restricts PLA’s processability in operations involving high stretch
rates, such as film blowing, thermoforming, and foaming. The last objective of this thesis is to
improve crystallinity and melt strength of biopolymers through reactive processing without
using any nucleating agents.
1.2 Thesis organization
This thesis contains seven chapters. Chapter 1 gives an introduction of biopolymers specifically
PHAs and PLA, their attributes and limitations. The chapter also defines the scope of the
proposed research. Chapter 2 summarizes the literature describing the various approaches that
have been followed to address the limitations of PLA and PHAs. Relevant work is examined
critically. Chapter 3 discusses the characterization of a series of MCL PHAs including true
molecular weight, melt viscosity, thermal stability, glass transition temperature (Tg), melting
temperature and crystallinity. Chapter 4 describes impact modification of brittle PHB, using a
MCL PHA (i.e. polyhydroxyoctanoate (PHO)). Chapter 5 describes the preparation of branched
PLA with improved strain hardening and crystallinity by reactive modification in the melt state,
using a peroxide and a multi-functional coagent. The performance of coagent-modified PLA is
5
compared with PLA modified with a multi-functional epoxide styrene-acrylic oligomeric chain
extender, containing glycidyl methacrylate (GMA) functions. Chapter 6 describes impact
modification of PLA, and further utilizes the reactive modification approach described in
chapter 5, to prepare reactively modified PHO and PLA/PHO blends to achieve a good balance
of mechanical and melt-state properties. Chapter 7 summarizes the overall outcomes and
achievements of the thesis with recommendations for future work.
6
Chapter 2 Literature Review
2.1 Polyhydroxyalkanoates
Polyhydroxyalkanoates (PHAs) are linear polyesters that were first discovered in late 1920s, by
Lemoigne who produced them using the Bacillus megaterium bacteria [7]. The chemical
structure of PHA is shown in scheme 2.1 [8].
Scheme 2.1 Chemical structure of polyhydroxyalkanoate [8]
PHA can be divided into three categories based on number of carbons in side chain (the R group
in Scheme 2.1): 1) Short-chain-length (SCL) PHA, 2) medium-chain-length (MCL) PHA and 3)
hybrid (mix of SCL PHA and MCL PHA). SCL PHA contains 0-2 carbons whereas MCL PHA
contains 3- 11 carbons in their backbone [9]. They can be found as homopolymers or co-
polymers. PHAs can have thermoplastic or elastomeric properties, depending upon their
composition. SCL PHAs behave like typical brittle thermoplastics, while MCL PHAs are
elastomeric. Melting points of PHAs lie between 40oC - 180oC. PHAs can be processed by
conventional polymer processing techniques.
Poly-3-hydroxybutyrate (PHB), a SCL PHA, is the most widely studied PHA. It is a brittle polymer
and cannot be used without impact modification. The high crystallinity and low crystallization
rate of PHB leads to embrittlement and an ageing effect of the polymer [10]. PHB is not
7
thermally stable [11], resulting in narrow processing windows [12]. The brittleness of PHB can
be counteracted by adding impact modifiers or by using copolymers of the same polymer
family, for example poly (3-hydroxybutyrate-co-3-hydroxyvalerate), P (3HB-co-3HV), which is
more ductile and flexible than PHB. Increase in the 3-hydroxyvalerate (3HV) content improves
flexibility with a decrease in melting temperature, tensile strength, modulus of elasticity and
crystallinity. P (3HB-co-3HV) was commercialized in the 1980s by Imperial Chemical Industries
Inc. (ICI) in the UK under the trade name Biopol®.
Medium-chain-length (MCL) PHA
SCL PHAs can be produced using a wide variety of bacteria under nutrient deprivation as an
intracellular energy reserve [13]. On the other hand, very few microorganisms can be used to
produce MCL PHA [14]. Accumulation of MCL PHA is restricted to Pseudomonas rRNA homology
group I, like Pseudomonas aeruginosa, P. chlororaphis, P. putida, P. syringae and some P.
fluorescens [14,15]. P. putida strains (formerly P. oleovorans GPo1) are used widely to produce
MCL PHA. They have the ability to use alkanes, such as octane for synthesis of MCL PHA due to
their octane (OCT) plasmid [13]. Alkanoates such as octanoate are a common carbon source to
make MCL PHAs [16].
The properties of MCL PHAs depend on the polymer composition. The melting point of
polyhydroxyoctanoate (PHO), a MCL PHA, is 61 °C with crystallinity of 30 % [17]. The glass
transition temperature (Tg) is -35 °C. It is elastomeric in nature, having an extension at break
value between 300-450 % and tensile strength between 6 and 10 MPa. The density of PHO is 1
g.cm-3 [17]. Based on its properties listed above, MCL PHA is a potential bio-sourced,
biodegradable impact modifier for brittle biopolymers.
8
Additives and chemical modification techniques that are commonly used to enhance the
performance of PHAs and broaden its processing window are outlined below.
2.1.1 Impact modification of PHB
Blending brittle polymers with a ductile polymer, which forms a secondary phase, is the easiest
way to impart flexibility in polymers. Polymer blends can either be miscible or immiscible.
Immiscible blends that are compatible yield properties better than the parent polymers
(synergistic effect) [18]. In the case of incompatible blends, compatibilizers can be used to
enhance the blend properties. Some of the PHA-based blends that have been investigated are
summarized below.
Parulekar et al. [19] used natural rubber to improve the ductility of PHB. They found that PHB-
natural rubber blends were not compatible and there was a substantial viscosity mismatch
between the two polymers. They used maleated polybutadiene with high graft content and low
molecular weight to compatibilize the system. They concluded that epoxidized (25%
epoxidation) natural rubber yielded better ductility compared to the non-functionalized natural
rubber. A maleated polybutadiene- PHB – natural rubber (10:60:30) system showed 440%
improvement in impact strength of PHB.
Block copolymers of polycaprolactone (PCL) and PHB have been used to compatibilize
immiscible PCL/PHB blends. Optimization of the composition of the main blend components
and compatibilizers is crucial to get substantial improvements in properties. 25 wt.% PCL and 5
wt.% of PCL-PHB block copolymers had only an elongation of 29 %, which increased
dramatically to 855 % when there was 45 wt.% of PCL and 10 wt.% of PCL-PHB block copolymer
[21]. Furthermore a 50:50 blend of PHB with a block copolymer of atactic poly ((R, S) -3-
9
hydroxybutyrate) and poly (ethylene glycol) resulted in increased elongation at break from 5 %
to 90% with decrease in modulus and tensile strength [22]. An immiscible blend of PHB and
acrylonitrile-g-(ethylene-co-propylene-co-diene)-g-styrene has also been studied and is
composed of four phases of poly (ethylene-co-propylene-co-diene), poly (styrene-co-
acrylonitrile), amorphous PHB and crystalline PHB. The blend exhibits 190% improvement in
impact resistance of pristine PHB [23].
2.1.2 Plasticization
The ductility of PHB can be improved by adding low molecular weight plasticizers. Generally
plasticizers decrease the intermolecular forces between polymer chains, thus increasing chain
mobility. Plasticizers for PHB include 1) high boiling esters of polybasic acids such as phthalates,
isophthalates, citrates, fumarates, glutamate, phosphates or phosphites 2) high boiling esters
and part esters of polyhydric alcohols mainly glycols, polyglycols and glycerol 3) aromatic
sulphonamides [24] and 4) a few high molecular weight polymers.
Ceccorulli et al. studied the effect of plasticization on the Tg of PHB [25]. They used a
biodegradable plasticizer, di n-butyl phthalate (DBP) to improve the ductility of the PHB. 30
wt.% of DBP was sufficient to lower the Tg of PHB from 6 °C to -40 °C. The data is in agreement
with Riande et al. [26] giving evidence of the existence of two concomitant phenomena, Tg
depression of the polymer due to the plasticizing effect with up to 40 wt.% of plasticizer, and a
small increase of the Tg of the plasticizer, due to hindrance in its mobility due to the dissolved
polymer molecule, above 40 wt.% of the plasticizer. Plasticization did not affect the ability of
PHB to crystallize. Increase in the plasticizer content decreases the crystallization temperature
10
as a result of reduced Tg that provides enough mobility to macromolecules for rearrangement
and crystallization.
Other plasticizers include glycerol triacetate (GTA), glycerol tributyrate (GTB), and glycerol
monostearate (GMS) [27]. Some high molecular weight polymers can also act as plasticizers. For
example, addition of PEO, which has a Tg of -59 oC lowers the Tg and crystallization temperature
of PHB [20].
2.1.3 Nucleation
One of the main drawbacks that have hindered widespread implementation of biopolyesters,
including PHAs, is their low crystallization rates, which result in poor and inconsistent
mechanical properties and processability. The time that these materials take to crystallize from
the melt can be a few hours, which is too long compared to most conventional polymers. This
makes them practically impossible to melt process in a cost-effective way using conventional
processing techniques. Improvements in the crystallization behavior are necessary to obtain
biopolymer-based formulations with good processability in polymer processing operations,
such as extrusion, injection molding, and film processing.
Nucleating agents are commonly used to overcome slow crystallization rates and to increase
the overall crystallinity, while maintaining small crystal sizes to achieve improved mechanical
properties. Some of the common nucleating agents used in industry include talc, mica, calcium
carbonate, chalk, and boron nitride. Environmentally friendly nucleating agents include
saccharin, and phthalimide but they are not as effective as the conventional nucleating agents,
such as boron nitride. They are both soluble in the melt and crystallize when solubility exceeds
the relevant threshold [28].
11
High molecular weight polymers like poly vinyl alcohol (PVA) can also be used to nucleate PHB.
PVA is a biodegradable, biocompatible and water soluble polymer with high crystallinity. A lot
of research has been done on blends of PHB and PVA, which form a miscible blend with lower
crystallinity. Alata et al. [29] showed that PVA particles can act as nucleating agents for PHB,
and enhance the rate of crystallization of PHB. PVA has a melting point of ~ 225 °C, so blends
processed at 190 °C ensure melting of PHB while PVA particles are still in the solid state. The
nucleating performance of PVA is equivalent to that of talc and provides a complete bio and
environment friendly material. Nano clays have also been used as nucleating agents for PHB.
Addition of montmorillonite (MMT) clay increased the crystallization temperature by 29 °C [30].
In search of bio-sourced alternatives, which are considered more sustainable, cellulose, which is
a naturally available crystalline material derived from wood (wood contains 40-50% cellulose
[31]), has been suggested as a potential candidate. Cellulose has been proven to act as a
nucleating agent for polymers like polypropylene. It reduces the size of the spherulites, while
increasing the overall crystallinity and inducing trans-crystallinity [32,33]. Most of the work
reported in this area has been done on microcrystalline cellulose. With recent advances in
nanotechnology, it has been shown that high-aspect ratio nano-sized fillers are advantageous
compared to their micron sized counterparts. Desired property improvements can be obtained
at a fraction of the loading, while loss of ductility is minimized. Nano-crystalline cellulose (NCC)
is a completely bioderived and biodegradable material, derived from cellulose through acid
hydrolysis. NCC has been the subject of many recent research initiatives, including partnerships
with the Canadian government [34]. In addition to being a promising candidate as a nucleating
agent for PHAs, it can also offer substantial reinforcement given its high strength [35].
12
2.1.4 Chain extension
Chain extension is used widely to increase the molecular weight of polyesters. Diisocyanate
forms urethane and amide linkages through a reaction with hydroxyl and carboxyl functional
group, resulting in significant increase in molecular weight. Hexamethylene diisocyanate (HDI)
has been used as a chain extender for PHO [4]. Scheme 2.2 illustrates the reaction of hydroxyl
and carboxyl functional groups of polyester with isocyanate forming urethane, amide and
allophanate linkages. This reaction increased the weight average molecular weight of PHO by
275 % and its number average molecular weight by 314 % [4].
Scheme 2.2 Reaction of isocyanate with hydroxyl and carboxyl functional groups, adapted from
Lee et al. 2009 [4]
2.2 Polylactic acid (PLA)
Poly (lactic acid) (PLA) (scheme 2.3) is an aliphatic polyester derived from renewable resources.
13
Scheme 2.3 Chemical structure of polylactic acid [36]
The properties of PLA depend significantly upon its molecular weight and the stereochemical
makeup of the backbone, which is controlled by polymerization with D-lactide, L-lactide, or D,L-
lactide, to form random or block stereocopolymers [37]. Some of the general properties of PLA
are summarized in Table 2.1
Table 2.1 Properties of PLA [37]
PLA
Density (Kg/m3) 1.26
Tensile strength (MPa) 59
Elastic modulus (GPa) 3.8
Elongation at break (%) 4-7
Notched izod (J/m) 26
Heat deflection temperature (°C) 55
Some of the limitations of PLA are quite similar to PHB, for example brittleness and low
crystallization rate. PLA has a very narrow processing window, because of the lack of melt
strength and its slow crystallization rates. Its poor engineering properties, including impact
strength and heat resistance have mainly confined its applications to food packaging [38], as
14
well as biomedical applications, such as drug delivery, where biocompatibility and
biodegradability are desired. Extensive research has been done on PLA to address these
problems. Various approaches to address the challenges of PLA are discussed below.
2.2.1 Impact modification
Blending with ductile polymers or elastomers is commonly used to counteract the brittleness of
PLA. Thermoplastic polyester elastomer (TPEE) has been used as an impact modifier for PLA.
Forming an immiscible blend with two phase morphology 4, 4-Methylenebis(phenylisocyanate)
acts as a compatibilizer increasing the interfacial adhesion between PLA and TPEE to give 340 %
increase in elongation at break while also maintaining modulus and tensile strength [39].
MCL PHA has been tested as an impact modifier for PLA. Immiscible blends of MCL PHA and
PLA, produced by solution mixing, exhibited substantial improvement in impact strength and
decreased tensile strength compared to the neat PLA [9]. Epoxy functionalized MCL PHA gave
further improvement in mechanical properties. The epoxy groups react with the hydroxyl group
of PLA, thus increasing the interfacial interaction and improving the blend morphology. Chain
extension reduced the viscosity mismatch between PLA and PHO, so that eventually the melt
viscosity of PHO and PLA blends was higher than that of blends without chain extension [4].
2.2.2 Plasticization
Polyethylene glycol (PEG) is often used to plasticize PLA. The system has improved elongation at
break with limited impact strength. The PEG based polyester, poly (polyethylene glycol-co-citric
acid) (PEGCA) forms a partially miscible blend with PLA. Addition of PEGCA diminishes the Tg of
PLA. At 15 wt.% it gives 242 % elongation at break with impact strength as high as 103 J/m [40].
15
Adipates like (bis (2-ethylhexyl) adipate and glyceryl triacetate) and polymeric adipates exhibit
excellent plasticization of PLA composites. They decrease Tg of PLA by 20 °C. 10 wt. % of
plasticizer gives four times higher impact properties for PLA composite containing 40 wt. % of
stable β -anhydrite. Adipates also help improve the dispersion of fillers and results in better
tensile strength of composites [41].
2.2.3 Nucleation
Nucleating agents are used to address the slow crystallization rate of PLA. They increase
nucleating density and thus increase crystallization rate. Talc exhibits effective nucleation in
PLA, whereas other nucleating agents like calcium lactate and sodium stearate had little or no
nucleating effect on PLA [42]. Talc contents of 1 wt.% significantly accelerate the crystallization
process of the PLA matrix. The maximum crystallization rate was observed at an annealing
temperature of 100 °C [43]. NCC has also been assessed in PLA. It has been shown that silylated
cellulose nanocrystals (SNCC) exhibits enhanced nucleating efficiency. Addition of 1 wt.% SNCC
resulted in substantial increase in crystallization rate. Increased crystallinity resulted in
improvements of tensile strength and modulus [44].
2.2.4 Conditioning
Annealing is another approach to toughen PLA. The annealing temperature and annealing time
have a significant effect on crystallization. As annealing time and temperature increase, the
impact strength increases due to smaller spherulite size and a larger amount of the metastable
phase. Quenching and subsequently annealing at an appropriate temperature (~90 °C) is crucial
for PLA toughening [44].
16
2.2.5 Chain extension
As most polyesters, PLA has poor melt strength which restricts its use in processes involving
extensional flow, such as film processing, thermoforming, and foaming. The processing window
of PLA can be broadened by chain extension. Chain extenders such as isocyanate, glycidol,
peroxides, and epoxy based styrene-acrylic oligomers have been used to increase the molecular
weight of PLA, to improve its thermal stability and to increase its melt strength by introducing
strain hardening [45], as explained in detail below.
2.2.6 Epoxy based chain extension
2.2.6.1 Glycidol
Deenadayalan et al. [46] achieved chain extension of PLA by reactive extrusion in the presence
of glycidol. Chain extension was initiated by reaction of the carboxyl and the hydroxyl of PLA
end groups with glycidol. The carboxylic end group of Poly(L-Lactic Acid) (PLLA) reacts with the
primary hydroxyl end groups of glycidol and forms end-capped linear PLLA (scheme 2.4).
Scheme 2.4 Reaction between PLLA and glycidol adapted from Deenadayalan et al. 2009 [46]
17
A hydroxyl end group from another PLLA reacts with glycidol and initiates chain extension with
formation of a pendant hydroxyl group that can react further with another end-capped PLLA to
form a branch structure. Using this reaction, the resulting increase in molecular weight was
more pronounced when low molecular weight as compared to high molecular weight PLA was
used; this was attributed to the higher concentration of end groups in low molecular weight
PLA, resulting in improved chain extension. The modified PLA showed higher Tg and melt
temperature.
2.2.6.2 Multi-functional epoxy based chain extenders
Joncryl®, a multi-functional epoxide styrene-acrylic oligomeric chain extender, containing
glycidyl methacrylate (GMA) functions (scheme 2.5) has been used successfully as a chain
extender for polyesters, such as PLA [47]. It has following physical characteristics: Tg – 55 °C,
epoxy equivalent weight – 285 g/mol.
Scheme 2.5 General structure of Joncryl® styrene – acrylic multi-functional oligomeric chain
extender, where R1 – R5 are H, CH3, a higher alkyl group, or combinations of them, R6 is an
alkyl group, and X, Y and Z are each between 1 and 20, adapted from Villalobos et al.2004 [48]
18
GMA contains epoxy groups, which participate in reactions with hydroxyl and carboxyl groups
[49]. In polyesters, glycidyl esterification of carboxylic acid leads to hydroxyl end group
etherification. Reaction of hydroxyl end group etherification competes with etherification of
secondary hydroxyl group and main chain trans-esterification. The epoxy ring opening reaction
results in covalent bonds via hydroxyl side group formation [50]. The reaction mechanism
proposed by Al-Itry is depicted in scheme 2.6 [51].
Scheme 2.6 Mechanism of PLA reaction with GMA, adapted from Al-Itry et al. 2012 [51]
2.2.7 Peroxide-mediated cross-linking
Peroxide curing is one of the oldest technologies to introduce branching and/or cross-linking in
polymers. It is used widely in elastomers and polyolefin technology. A general scheme of
peroxide cross-linking is shown in section 2.2.8. The peroxide undergoes hemolytic
decomposition upon heating to generate free radicals. Free radicals abstract hydrogen from the
polymer chain creating polymer radicals and decomposition products. The polymer radicals can
Degradation
GMA
GMA
Acid End GroupVinylic terminated ester
PLA
19
react with each other to form a C-C crosslinks. A competitive degradation reaction can also take
place through β-chain session.
In PLA, peroxide-mediated chain extension has been achieved by reactive extrusion using
lauroyl peroxide [52] dicumyl peroxide [53] and di-tertiary alkyl peroxide [54] with limited
success. The choice of peroxide typically depends on the peroxide half-life time and the
polymer processing temperature.
2.2.8 Multifunctional coagents
Multifunctional coagents are used widely to form branched polyolefins [55,56]. Parent et al.
[57] studied triallyl trimesate (TAM), trimethylolpropane triacrylate (TMPTA) and triallyl
phosphate (TAP) coagents (Scheme 2.7) in polypropylene. Depending on the type of coagent,
different molecular weight and branching distributions may be obtained.
Scheme 2.7 Chemical structure of coagents, adapted from Parent et al. 2008 [56]
A typical peroxide assisted coagent modification scheme is illustrated below.
20
Initiation
ROOR 2 RO•
RO• + P-H ROH – P•
Propagation
P• + M. → P-M•
P-M• + P-H → P-MH + P•
Chain transfer
P• + P-H P-H + P•
Fragmentation -
P• P= + P•
Termination
P• + P• → P-P or P-H + P=
More specifically, the TAM activation reaction is discussed by Parent et al. [56].
21
Chapter 3 Determination of Mark-Houwink parameters and absolute
molecular weight of medium-chain-length poly(3-
hydroxyalkanoates)*
3.1 Introduction
Current global production of commodity plastics consumes up to 270 million tons of petroleum
annually [58]. Despite recycling efforts, as much as 50% of the 100 million tons of plastic
produced annually may end up in landfill sites. To address environmental and resource issues,
"bio-based plastics” that are both biodegradable and made from renewable resources are being
developed for a variety of applications. New applications and innovations are anticipated in the
automotive, electrical/electronic, medical and packaging industries. Bioplastics include bio-
polyethylene, polylactic acids (PLAs), polyhydroxyalkanoates (PHAs), and starch-based
materials. Forecast analysts including Technavio and Helmut Kaiser Consultancy predict that the
bioplastics market will experience significant growth (33.9-41% compound annual growth rate
(CAGR)) from 2010 to 2015 [59].
As mentioned in section 2.1, the physical properties and solubilities of PHA are greatly affected
by the length of the side group R (Scheme 2.1). Poly(3-hydroxybutyrate) (PHB), the most
common short-chain-length (SCL) PHA, is stiff and brittle, whereas medium-chain-length (MCL)
PHAs are more elastic and flexible. MCL PHAs are biodegradable materials with low crystallinity,
low glass transition temperature (Tg), high elongation, and low tensile strength [60], making
them good candidates for applications where elastomeric properties are needed, such as for
*A version of this chapter has been published. Nerkar, M., Ramsay, J.A., Ramsay, B.A., Kontopoulou, M., Hutchinson, R.A. Journal of Polymers and the Environment 2013 (21): 24-29
22
impact modification of brittle biopolymers [47,61]. MCL PHAs are the only thermoplastics
elastomers that have 100 % renewable content and are biodegradable.
Given the influence on many engineering properties such as strength, stiffness, toughness,
elasticity, viscosity and thermal transitions, it is important to have access to reliable molecular
weight data for PHAs, to assess the effect of synthesis conditions on their properties and
evaluate the potential of these biopolymers in engineering applications.
Of the various PHA materials, PHB is by far the most investigated and characterized polymer. As
PHB is not soluble in tetrahydrofuran (THF), molecular weight (MW) characterization is done in
more aggressive solvents such as chloroform, 2,2,2-trifluoroethanol, and ethylene dichloride.
Akita et al. [62] and Marchessault et al. [63] employed light scattering and osmometry to obtain
absolute molecular weight (MW) data and to determine the Mark-Houwink (MH) calibration
constants required to estimate PHB molecular weight distributions (MWD) from size exclusion
chromatography (SEC) analysis coupled with a single detector calibrated using polystyrene (PS)
standards. Ubbelohde type capillary viscometry and rotational viscometry have also been used
to determine intrinsic viscosity data [62]. Miyaki et al. [64] and Cornibert et al.[65] covered a
broader molecular weight range using osmometry. These efforts allow the estimation of
absolute MW values from single-detector SEC analysis: a value of 1.0105 Da as measured by PS
calibration is transformed to 0.6105 Da for PHB in chloroform using the MH constants
published by Akita et al., a shift of 35%. PHB molecular weight is typically in the order of 1106
Da in native granules but may decrease if exposed to depolymerases, base, oxidizing agents
such as hypochlorite [66] or chlorinated solvents [67].
23
However, only limited MW data is available in the literature for MCL PHAs [60] and MCL SCL
PHA materials [68], as measured by SEC analysis in THF eluent, and MW values are typically
reported relative to PS calibrations. Absolute values of molecular weight for MCL PHAs are not
reported in the literature to date, as the necessary SEC calibrations had not been performed.
This information is required to better relate polymer structure to the processing properties and
the degradation kinetics of these emerging biomaterials, such as those reported by Daly et
al.[69] for poly (3-hydroxybutyrate-co- 3-hydroxyhexanoate).
In this chapter, absolute MW averages have been determined for four different MCL PHA
copolymers from MWDs measured using multi-detector SEC, and MH parameters were
estimated for the first time for these kinds of polymers in THF. Furthermore the melt viscosity is
reported as a function of molecular weight.
3.2 Experimental
3.2.1 Materials
The four samples were copolymers with (3-hydroxyoctanoate) (PHO), (3-hydroxynonanoate)
(PHN) and (3-hydroxydodecanoate) (PHDD) moieties produced from renewable starting
materials including sugar and vegetable oil. PHO contained 98 mol % 3-hydroxyoctanoate and 2
mol % 3-hydroxyhexanoate. PHN 90 and PHN 70 contained 3-hydroxynonanoate and 3-
hydroxyheptanoate at mole ratios of 90/10 and 70/30, respectively. PHDD contained 40 mol %
3-hydroxydodecanoate 39 mol % 3-hydroxydecanoate, 19 mol % 3-hydroxyoctanoate and 2 mol
% 3-hydroxyhexanoate.
24
PHN 90 and PHO were produced in chemostat culture with addition of acrylic acid to inhibit β-
oxidation [70]. PHN 70 and PHDD were produced in fed-batch culture [71]. The polymer
samples were extracted from washed lyophilized biomass by Soxhlet extraction in acetone
followed by precipitation in cold methanol [72], except for PHDD which was extracted in
chloroform [67]. Cellular PHA content and composition were determined by gas
chromatography as described by Sun et al. [71].
3.2.2 Methods
Samples were prepared for SEC analysis by dissolving 10 mg of polymer in 1 mL of distilled THF
overnight to ensure complete dissolution, then passed through a 0.2 µm nylon filter. Polymer
molecular weight distributions were measured using two different instruments to check
consistency and reliability of the data. The first, a Viscotek 270max separation module with
triple detection by differential refractive index (DRI), viscosity (IV) and light scattering (low
angle LALS and right angle RALS), was maintained at 40 °C and contained two porous
PolyAnalytik columns in series with an exclusion molecular weight limit of 20106 Da. Distilled
THF was used as the eluent at a flow rate of 1 mL/min. The MWDs were calculated using two
methodologies. First, the results from the triple detector train and Viscotek Omnisec
software were used to determine polymer MWDs and MW averages using the values of the
refractive index (dn/dc) determined offline, using a Wyatt Optilab DSP refractometer as
described below. The triple detection mode also yielded estimates for the polymer Mark-
Houwink (MH) parameters, determined directly from the curve generated by the output from
the IV and LS detectors. These MH parameters provided the means for a second analysis of the
output data using the principle of universal calibration and the output from the DRI detector,
25
with a calibration curve constructed with narrow molecular weight polystyrene standards
ranging from 6910 to 3.3106 Da.
Samples were also characterized on a second SEC instrument consisting of a Waters 2960
separation module coupled with a Waters 410 differential refractometer (DRI) and a Wyatt
Instruments Dawn EOS 690 nm laser photometer multiangle light scattering (LS) detector. THF
was used as eluent at a flow rate of 1 mL/min through the four Styragel columns (HR 0.5, 1, 3,
4), maintained at 35 °C. The DRI detector was calibrated by 10 narrow polydispersity
polystyrene standards in a broad MW range (870-3.55105 Da), and the LS detector was
calibrated by toluene, as recommended by the manufacturer. MWDs from this instrument were
again calculated using two methods, using the output from the DRI detector and the MH
parameters determined with the Viscotek setup, and by processing the data from the LS
detector using the Wyatt Astra software and the refractive index (dn/dc) of the polymer in THF,
according to standard procedures. The refractive index values were measured by a Wyatt
Optilab DSP refractometer at 35 °C and 690 nm calibrated with sodium chloride. Five samples of
3–18 mgmL–1 were prepared in THF for each polymer and injected sequentially to construct a
curve with slope dn/dc [73]. The values for PHN 90 and PHO were 0.06010.0002 mL/g and
0.06030.0003 mL/g, respectively.
DSC experiments were performed using a Q100 DSC from TA Instruments, under dry nitrogen.
Since MCL PHAs crystallize slowly, the samples were preconditioned to eliminate their thermal
history as follows: the polymer was heated at 100 °C for 10 min in a convection oven, and then
kept at room temperature for two weeks before characterization. Samples weighing 10-12 mg
were sealed in aluminum hermetic pans, equilibrated at -70 °C and kept isothermally for 5 min.
26
Afterwards they were heated to 100 °C at a rate of 5 °C/min and held isothermally for 3 min
before cooling to -70 °C at a rate of 5 °C/min. As MCL PHAs did not crystallize during second
heating cycle data from first heating cycle was used for differentiation. The samples were finally
reheated to 100 °C at a rate of 5 °C/min. The % crystallinity of the polymers, Xc, was estimated
using equation (3.1).
100H
HX
100
mc
(3.1)
where, ΔHm is the enthalpy of fusion and ΔH100 is the theoretical fusion enthalpy of a 100%
crystalline polymer, which is 146 J/g [74].
Rheological characterization was carried out in the constant rate mode under nitrogen blanket
using a ViscoTech rheometer by Reologica, equipped with a cone and plate fixture having 25
mm diameter at 120 °C. All MCL PHAs had Newtonian behavior.
3.3 Results and Discussion
3.3.1 Molecular weight determination
Two different instruments, a triple detector Viscotec 270 max and a dual detector
Waters/Wyatt, were used to characterize the polymer samples, as detailed in the Experimental
Section. MWDs and molecular weight averages were calculated from each instrument using
both multidetector and single detector (DRI) with universal calibration) output.
Log-log plots of polymer intrinsic viscosity ([] in dL/g) vs. MW were constructed using the
triple detector output from the Viscotec SEC; experimental results for the three PHO replicates
are shown in Figure 3.1a. These curves were calculated using the dn/dc value of 0.0602 mLg–1
27
determined to be independent of the MCL PHA composition in THF at 35 °C. Linear regression
was used to estimate the MH calibration parameters and for all four samples, as summarized in
Table 3.1. Averages and standard deviations of three measurements are reported. The four sets
of calibration parameters represent a very similar [] vs MW behavior for the copolymer
samples (Figure 3.1b). Thus, the values were averaged to provide an estimate of a “universal”
set of MH parameters that may be applied to all MCL PHAs, independent of composition. The
last column in Table 3.1 reports MW values calculated for a PS equivalent MW of 105 Da using
the four individual pairs of MH parameters as well as that calculated according to the
“universal” MH parameters. The difference between the various estimates is less than 15%,
which is the typical error reported for MW measurements by SEC. Moreover, the calculated
values are within 10% of 105 Da, the PS equivalent MW.
All of the copolymers were characterized at least three times on three different days to check
reproducibility of the data. The MWD data of PHO (three replicates run on two instruments
analyzed using multidetector and universal calibration) are shown in Figure 3.2. The peak
positions of the three replicates are tightly grouped at 105 Da, independent of the SEC setup or
analysis method chosen. This agreement, also observed for the other samples, demonstrates
the validity of the polymer refractive index (dn/dc) and MH calibration parameters determined
in this work for MCL PHA materials.
Before drawing generalized conclusions, it is useful to examine the number-average (Mn) and
weight-average (Mw) molecular weight values as well as the dispersity index (PDI= Mw/Mn) for
the four MCL PHA samples. Only the output from the Viscotek instrument was used, as the
column set used in the Waters/Wyatt instrument is designed for analysis of lower MW samples
28
and the calibration for the DRI detector only extends to 3.5105 Da. Thus, the polymer MWDs
are cut off at higher molecular weights, most clearly seen in Figure 3.2a with the DRI detector.
Figure 3.1 a) Experimental Mark-Houwink data for PHO (three replicates), b) best fit Mark-
Houwink relationships for all copolymer samples.
(The Wyatt ASTRA software used for processing of the LS detector provides an estimate of the
complete MWD; however, separation of the higher MW polymer will be incomplete due to the
column set used.) MW averages and standard deviations are reported in Table 3.2 as estimated
a)
b)
29
by the full Viscotek triple detector (TD) analysis, as well as using only the output from the DRI
detector coupled with universal calibration (DRI/UC). The universal calibration procedure is
done using the sample-specific MH parameters reported in Table 3.1, as well as the averaged
“universal” MH sample-specific MH parameters reported in Table 3.1, as well as the averaged
“universal” MH parameters. Furthermore, the MW averages are also reported according to DRI
analysis with PS calibration.
Table 3.1 Mark Houwink calibration constants from best fit to triple detector analysis of MCL
PHA samples in THF ([Ƞ]=K(MW)a).
a Calculated for PS equivalent MW of 105 Da
As summarized in Table 3.2, the MWDs of the four samples have similar PDI values of 1.8-2.3.
(The values obtained by TD-SEC are lower, as is often observed in MWDs measured by light-
scattering techniques). However, as also seen in the MWDs plotted as Figure 3.3, the absolute
a Log(K/(dL·g-1)) MW (kDa)a
PS [75] 0.716 -3.943 100.0
PHO 0.701 (0.013) -3.77 (0.07) 88.1
PHN 90 0.663 (0.020) -3.62 (0.11) 92.7
PHN 70 0.701 (0.031) -3.86 (0.16) 99.8
PHDD 0.691 (0.013) -3.86 (0.07) 106.0
Universal PHA 0.689 -3.78 96.2
30
Mw averages vary from 3.0-3.4104 Da for the PHN-70 sample to 1.4-1.7105 Da for PHO. These
values, controlled by the synthesis conditions for the different copolymers, are within the range
Figure 3.2 Molecular weight distributions for PHO (three replicates) as measured by a)
Waters/Wyatt SEC, light-scattering detector; b) Waters/Wyatt SEC, universal calibration; c)
Viscotek SEC, triple detection; and d) Viscotek –SEC, universal calibration.
a) b)
c) d)
31
Table 3.2 Number-average (Mn), weight-average (Mw), and dispersity (PDI) values determined from SEC analysis of MCL PHA
samples.
Values are reported according to polystyrene (PS) calibration, universal calibration (UC) with individual polymer MH parameters
from Table 3.1, UC with “universal” parameters for MCL PHA, and triple detector (TD) analysis. The average of three samples is
reported.
PS Calibration UC, individual MH parameters UC, averaged MH parameters TD Analysis
Mn (kDa) Mw (kDa) PDI Mn (kDa) Mw (kDa) PDI Mn (kDa) Mw (kDa) PDI Mn (kDa) Mw (kDa) PDI
PHO 73.74.5 15016 2.020.11 61.65.7 140.715.3 2.280.22 67.26.3 155.217 2.310.22 98.2±6.7 172±17 1.75±0.10
PHN90 52.71.6 98.32.6 1.87010 44.92.0 92.62.4 2.060.14 47.42.1 95.62.4 2.030.14 49.8±1.8 89.7±7.0 1.80±0.14
PHN70 15.40.7 33.60.5 2.180.11 13.60.8 31.10.4 2.290.14 12.90.7 29.90.4 2.320.14 18.2±3 31.9±2.3 1.77±0.17
PHDD 25.71.4 47.11.4 1.830.05 24.51.6 47.31.5 1.930.07 22.31.4 43.21.3 1.930.07 29.7±1.3 46.8±0.8 1.58±0.04
32
Figure 3.3 Molecular weight distributions of four MCL PHA polymers (Viscotek SEC, universal
calibration).
reported in the literature for MCL PHAs [60,68].The difference in Mw values calculated using TD-
SEC and DRI/UC methodologies is the largest for PHO, with the spread for the other three
(lower Mw) samples less than 10%. This very good agreement (Table 3.1) validates the adoption
of a “universal” set of MH parameters for MCL PHA samples analyzed in THF, and is
independent of copolymer composition.
3.3.2 Thermal and rheological characterization
The thermal properties of MCl PHAs are summarized in Table 3.3. All the materials were highly
amorphous, with PHO having the highest melting temperature and lowest crystallinity, whereas
PHDD has highest crystallinity.
33
Table 3.3 Thermal properties of MCL PHAs
TM (°C) Tg (°C) Crystallinity (%)
PHO 63 -36 15 PHN 990 60 -46 17 PHN 970 50 -47 16 PHDD 61 -40 21
The bulk rheology of the four samples was also measured. All samples demonstrated
Newtonian behavior. The zero shear viscosity scales with Mw with a slope of 1 at lower
molecular weights, and has a slope of 3.8 which is higher than the slope of 3.4 anticipated for
linear polymers at the higher molecular weights, as shown in Figure 3.4. This may be due to
experimental error. The plot suggests an entanglement molecular weight around 8104 Da
(where the two fitted lines intersect), which is significantly higher than most conventional
polymers, suggesting that these polymers may adopt folded helical conformations, similar to
what has been proposed for PHB [76].
Figure 3.4 Viscosity as a function of molecular weight
1
10
100
1.E+04 1.E+05 1.E+06
η ( P
a S
) a
t 1
20
°C
Mw (Da)
η~Mw
η~Mw3.8
34
3.4 Conclusion
This study provides the first determination of absolute MWDs and MW averages of MCL PHA
copolymers. Fortuitously (and unlike PHB in chloroform), the relationship between polymer
MW and intrinsic viscosity is very close to that determined through PS calibration. Thus,
previously reported MW data [67,77] for MCL PHAs relative to polystyrene calibration can be
considered, within experimental error, as absolute values. With these results, it will be possible
to more closely examine the relationship between MCL PHA synthesis conditions and polymer
MWs, and to better assess their processability using viscosity data. The latter will be of
significant benefit in product development.
35
Chapter 4 Melt compounded blends of short and medium-chain-length
poly-3-hydroxyalkanoates*
4.1 Introduction
Poly(3-hydroxyalkanoates) (PHAs) are microbially produced, biodegradable polymers derived
from renewable resources [6]. Poly-3-hydroxybutyrate (PHB), the most studied short-chain-
length (SCL) PHA, is a brittle polymer requiring modification to render it suitable for various
engineering applications. Incorporation of 3-hydroxyvalerate (3HV) to form a copolymer of
poly(3-hydroxybutyrate-co-3-hydroxyvalerate) (PHB-HV) results in improved ductility, impact
properties, and lower tensile strength, moduli, melting temperatures and crystallinity [78].
Blends of PHB with other polymers have been studied extensively [79,80]. Parulekar et al. [19]
achieved 440% improvement in the impact strength of PHB by adding a maleated
polybutadiene to PHB/natural rubber blends. Substantial improvements in the impact
resistance of PHB have been noted in immiscible blends of PHB with acrylonitrile-g-(ethylene-
co-propylene-co-diene)-g-styrene, comprising of four phases: poly (ethylene-co-propylene-co-
diene), poly (styrene-co-acrylonitrile), amorphous PHB and crystalline PHB [23].
Polycaprolactone (PCL)/PHB blends containing a PCL-PHB block copolymer as compatibilizer,
exhibited dramatic increases in the elongation at break [21]. Blends of PHB containing a block
copolymer of atactic poly((R, S)-3- hydroxybutyrate) and poly(ethylene glycol) resulted in
increased elongation at break and lower modulus and tensile strength [22].
PHB/poly(methyleneoxide)(POM) blends exhibited modest improvements in mechanical
*A version of this chapter has been published. Nerkar, M., Ramsay, J.A., Ramsay, B.A., Kontopoulou, M., Journal of Polymers and the Environment, 2014 (22): 236-243
36
properties [20]. Blends of PHB with polymers derived from renewable sources, such as
poly(lactic acid) (PLA), starch and chitosan have also been studied [80-82].
Medium-chain-length (MCL) PHAs, such as poly (3-hydroxyoctanoate) (PHO), have low
crystallinity and exhibit elastomeric properties. PHO has been used to impact modify PLA via
solution blending [9]. Epoxy functionalized MCL PHA resulted in further improvements in
mechanical properties. These materials are therefore promising biopolymers as impact
modifiers for PHB [60]. Dufresne et al. [83] noted a transition from elastomeric to brittle
properties upon addition of PHB to PHO using a solvent mixing technique. Martelli et al. [84]
noted 50% improvement in elongation at break of PHB-HV upon addition of MCL PHA using
solution casting. However melt compounding, which is more industrially relevant, has not been
investigated, in part due to the technical difficulties in producing sufficient quantities of MCL
PHAs.
The objective of this chapter is to prepare and characterize melt compounded blends of MCL
PHAs with PHB. In the previous chapter the properties of four different MCL PHAs with
predominantly 3-hydroxyoctanoate (PHO), 3-hydroxynonanoate (PHN) or 3-
hydroxydodecanoate (PHDD) content were reported. PHO had the highest melt viscosity and
molecular weight. It also has low crystallinity and glass transition temperature (Tg). Based on
these attributes PHO was chosen as the preferred candidate for impact modification in the
present chapter. The morphology, mechanical, thermal and rheological properties of PHB/PHO
blends are reported. Furthermore peroxide-initiated cross-linking of PHO is used to counteract
the viscosity mismatch between the two components.
37
4.2 Experimental
4.2.1 Materials
An unmodified, additive free PHB (grade T19), with a weight-average molecular weight of
1,416,000 Da with a dispersity of 7, was supplied by BIOMER, Krailling, Germany in the form of
powder. PHO, containing 98 mol % 3-hydroxyoctanoate and 2 mol % 3-hydroxyhexanoate, was
produced and characterized as described previously in chapter 3 [85]. Its weight average
molecular weight, determined by triple-detector size exclusion chromatography (SEC), was
172,000 Da with a dispersity of 1.75. Lauroyl peroxide (L-231) and ethyl ether anhydrous were
obtained from Elf Atochem and Sigma Aldrich respectively and were used as received.
4.2.2 Compounding
PHB and PHO were dried in a vacuum oven at 100 °C and at room temperature respectively, to
remove moisture. Blends containing 0-30 wt.% PHO were compounded in a DSM
microcompounder at 190 °C for 3 min at a screw speed of 100 rpm. The compounder was
operated under nitrogen blanket to limit polymer degradation. After compounding, the strands
were quenched in cold water before chopping into pellets.
4.2.3 PHO cross-linking
Weighed amounts of lauroyl peroxide (0.06-0.5 wt.%) were dissolved in anhydrous ethyl ether
and coated onto PHO pellets in a glass Petri dish. The coated pellets were placed in a vacuum
oven overnight at room temperature to remove the solvent. Cross-linking was conducted in a
Carver press at 155oC for 10 min to ensure complete reaction (the half-life time of the lauroyl
peroxide at 155oC is 0.8 min). Cross-linked PHO was chopped into small pieces and dry mixed
with PHB at appropriate ratios before feeding into compounder. For rheological
38
characterization, the cross-linked compression molded sheets were cut into 25 mm diameter
disc and used for testing as described below.
The gel content of the peroxide-cross-linked MCL PHA was measured by dissolving the material
in boiling tetrahydrofuran (THF) for 7 h. The polymer was sealed in stainless steel wire mesh
(120 mesh) according to ASTM D 2765. The material was left for 1 h under the fumehood and
subsequently dried overnight in a vacuum oven at room temperature. The % gel content was
calculated using equation (4.1).
100sample of weight Initial
sample of weight Finalcontent Gel (4.1)
4.2.4 Blend Characterization
4.2.4.1 Differential scanning calorimetry (DSC)
DSC experiments were performed using a Q100 DSC from TA Instruments, under dry nitrogen.
Since MCL PHAs crystallize slowly, the samples were preconditioned to eliminate their thermal
history as follows: the polymer was heated at 100 °C for 10 min in a convection oven, and then
kept at room temperature for two weeks before characterization. Samples weighing 10-12 mg
were sealed in aluminum hermetic pans, equilibrated at -70 °C and kept isothermally for 5 min.
Afterwards they were heated to 200 °C at a rate of 5 °C/min and held isothermally for 3 min
before cooling to -70 °C at a rate of 5 °C/min. The samples were finally reheated to 200 °C at a
rate of 5 °C/min. The % crystallinity of the polymers, Xc, was estimated using equation (4.2).
100H
HX
100
mc
(4.2)
where ΔHm is the enthalpy of fusion and ΔH100 is the theoretical fusion enthalpy of a 100%
crystalline polymer, which is 146 J/g [74].
39
4.2.4.2 Thermogravimetric analysis (TGA)
TGA was conducted on a Q500 TGA from TA Instruments under nitrogen atmosphere, using 6-8
mg samples. The weight loss was evaluated by heating to 800 °C at a heating rate of 20 °C/min.
Isothermal experiments were conducted at the compounding temperature of 190 °C for 20 min.
The temperature at which 5 wt.% degradation occurred was reported as the degradation onset
temperature.
4.2.4.3 Scanning electron microscopy
Blend morphologies were observed using a JEOL JSM-840 scanning electron microscope.
Samples were first hot-pressed at 200oC for 3 min, then immersed in liquid nitrogen for 3 min
before brittle fracture. The MCL PHA phase was etched in acetone overnight at room
temperature.
4.2.4.4 Rheology
Compression molded discs, 25 mm diameter and 2 mm thick, were prepared using a Carver
press. The linear viscoelastic properties were measured in the oscillatory mode using a stress
controlled rheometer, Visco Tech from Reologica, under nitrogen purge. Frequency sweeps
were conducted at 190 °C using a cone and plate fixture having 25 mm diameter and 2° angle,
at a frequency range between 1-100 rad/s. This range was chosen to limit the duration of the
experiment to 100 s, given the sensitivity of PHB to degradation. Time sweeps were conducted
at a frequency of 6.28 rad/s and temperatures ranging from 135 to 155 °C to obtain cure curves
for the cross-linked PHO formulations.
40
4.2.4.5 Mechanical properties
The compounded materials were pre-dried in a vacuum oven at room temperature overnight.
Specimens for mechanical property characterization were prepared by compression molding
using a Carver press under 5000 N force, at 200 °C and a residence time of 3 min, then
quenched in cold water. All specimens were conditioned at room temperature for 48 h after
compression molding, prior to mechanical testing. Tensile tests were conducted in accordance
with ASTM D638 using standard type V test specimens, with an Instron 3369 Universal tester, at
a cross head speed (CHS) of 5 mm/min. The average of five measurements is reported. Un-
notched Izod impact tests were conducted in accordance with ISO 180 using standard
specimens on a SATEC Instron machine and the average of five specimens are also reported.
4.3 Results and discussion
4.3.1 Thermal and rheological properties
Thermal degradation is an important concern when processing PHB. The isothermal TGA curves
(Figure 4.1) show that under nitrogen atmosphere PHB started to degrade at times longer than
10 min at the compounding temperature of 190 °C. Based on the TGA data, PHO had better
thermal stability than PHB and its addition to PHB improved the thermal stability of the blend.
The by-product of PHB degradation is mainly trans-2-butenoic acid [86], whereas the products
of PHO degradation would be a mixture of trans-2-octenoic and trans-2-hexenoic acids. The
presence of different decomposition products may affect the degradation kinetics of the
mixture.
41
Figure 4.1 TGA curves for PHO, PHB and the 85/15 PHB/PHO blend at 190 °C
Table 4.1 summarizes the degradation onset temperatures of the blends in non-isothermal
experiments. Addition of PHO gradually increased the degradation onset temperature of the
blends. The shift in temperature was as high as 20 °C for the blends containing 30 wt.% PHO.
Figure 4.2 shows the DSC thermograms of PHB and PHB/PHO blends. The first heating cycle of
the PHB showed a single melting peak at 181 °C and a corresponding crystallinity of 66%. A
double peak appeared during the second heating cycle (Figure 4.2a). The first melting peak at
the lower temperature (171 °C) corresponds to crystals formed at the crystallization
temperature and the second one at 177 °C is due to crystals that form during the heating cycle
[83,87,88]. The upper peak is generally attributed to the presence of more stable crystals that
94
95
96
97
98
99
100
101
0 5 10 15 20
We
igh
t (%
)
Time (min)
PHO
PHB
85/15 PHB/PHO
42
are favoured when unstable crystals melt and reorganize during the heating scan at slow
cooling rates, such as the ones employed in DSC [88].
Table 4.1 Crystallization temperature (TC), first and second melting peaks (TM1 and TM2
respectively), % crystallinity and degradation onset temperature for PHB and PHB/PHO blends.
PHB/PHO
(wt.%/ wt.%) TC (°C) TM1 (°C) TM2 (°C) Crystallinity (%)
Degradation
onset temp (°C)
100/0 101 171 177 64 254
95/5 97 171 177 58 261
90/10 91 173 178 55 261
85/15 96 172 178 53 264
80/20 87 172 177 53 273
70/30 78 169 177 45 277
The relative magnitude and position of these peaks generally depends on the heating rate [89],
and the blend composition [23]. In the present study, the high temperature endotherm became
more prominent as the PHO content increased, suggesting that crystals that formed during the
heating cycle were favored at higher PHO content. Moreover, as the amount of PHO was
increased in the blend, the PHB crystallization peak shifted to lower temperatures (Figure 4.2b),
suggesting that the PHB crystalline structure is affected in the presence of PHO.
43
Figure 4.2 a) DSC endotherm (2nd heating cycle) and b) DSC exotherm of PHB and PHB/PHO
blends
150 155 160 165 170 175 180 185
Heat
Flo
w (
W/g
)
Temperature ( C)
30 50 70 90 110 130 150
He
at
Flo
w (
W/g
)
Temperature ( C)
5 wt % PHO 15 wt % PHO
30 wt % PHO PHB
a)
b)
44
The PHO had very low crystallinity. A first heating cycle revealed a Tg of -36 °C, melting
temperature of 63 °C and crystallinity of 15 %. Given its very low crystallization rates, PHO did
not crystallize during the cooling cycle, therefore the second heating cycle was entirely
featureless and is not shown in Figure 4.2. Addition of PHO to PHB resulted in a gradual
decrease in the heat of fusion, resulting in decreased crystallinity, as shown in Table 4.1. This is
a common phenomenon when an elastomeric material is added to a semi-crystalline one, and
has also been noted when PHB was blended with amorphous polymers [81].
The linear viscoelastic properties of PHB and PHO are summarized in Figure 4.3. Both polymers
exhibited Newtonian behavior, with no shear thinning over the frequency range investigated, as
shown in Figure 4.3a. A large viscosity mismatch between the two polymers at the
compounding temperature was noted, with PHB being significantly more viscous. This is
obviously attributed to the differences in the molecular weight (172,000 Da and 1,416,000 Da
for PHO and PHB respectively). The viscosity ratio, ⁄ , (where ηd is the viscosity of the
dispersed phase and ηm is the viscosity of the matrix) is very low, about 0.03 and affects
negatively the dispersion of the dispersed phase within the matrix, as discussed in the following
section.
4.3.2 Morphology
The PHB/PHO blends had droplet-matrix morphology, typical of immiscible polymer blends, as
seen in the SEM images (Figure 4.4). The PHO domains were well dispersed within the PHB
matrix at low PHO content, but as the level of PHO was increased, the domain sizes became
larger and the morphology deteriorated. At 30 wt.% PHO, the coalescence of the dispersed
45
Figure 4.3 Rheological properties of PHB and PHO and effect of peroxide cross-linking on a)
complex viscosity, b) storage modulus and c) loss tangent, tan δ, measured at 190 °C
1
10
100
1000
10000
100000
1 10 100
Co
mp
lex
Vis
co
sit
y (
Pa
s)
Frequency (rad/s)
100
102
103
100 101 102
104
105
101
0.01
0.1
1
10
100
1000
10000
100000
1 10 100
Sto
rag
e M
od
ulu
s
(Pa
)
Frequency (rad/s)
102
103
104
101
100
10-1
10-2
100 101 102
105
0.01
0.1
1
10
100
1000
1 10 100
tan
δ
Frequency (rad/s)
PHB PHOPHO-0.06 wt % L-231 PHO-0.2 wt % L-231PHO-0.5 wt % L-231
10-2
10-1
100
101
102
103
100 101 102
a)
b)
c)
46
phase resulted in a very coarse structure. As explained above, the tendency for coalescence
may be attributed to the significant viscosity mismatch between the two blend components.
Better morphology was reported in solution blending of PLA and PHO having similar viscosities
[9]. However viscosity is not a factor during solution blending, therefore the results are not
directly comparable.
Figure 4.4 Scanning electron microscopy blends containing a) 5 wt.%, b) 10 wt.%, c) 15 wt.%, d)
20 wt.% and e) 30 wt.% of PHO at 2000x magnification
4.3.3 Mechanical properties
The tensile strain increased with the PHO content (Figure 4.5a). Considerable enhancement was
observed above 15 wt.% PHO, indicating improved flexibility of the blends. At 30 wt.%, there
was a decline in tensile strain. At this high PHO content, the PHO domains coalesced (Figure
4.4) and the deterioration in morphology caused the tensile strain to decrease. The un-notched
impact strength of the blends improved by only 50% when 20 wt.% PHO was added, but
47
increased by 150% with 30 wt.% PHO compared to pristine PHB. On the other hand, the tensile
stress and the Young’s moduli decreased, as shown in Figure 4.5b. The decrease in tensile stress
is typical of impact modification and can be justified by the decrease in crystallinity with the
addition of PHO in the blend and the introduction of a softer component in the blend.
Improvements in the strain at break, and a decrease in Young’s modulus and tensile strength
have been reported in PHB-HV/MCL PHA blends prepared by solution blending [84]. However in
these blends MCL PHA contents above 5 wt.% led to phase separation and a decrease in strain
at break. The results outlined above suggest that addition of PHO to PHB reduced the
crystallinity of the blend, and moderately increased impact and elongation, which were
counteracted by a decrease in the modulus. These are the expected results of impact
modification. The extent of impact modification however remains limited, due to the coarse
morphology, especially at high loadings which is attributed to the viscosity mismatch between
the blend components, attributed to the very low viscosity of PHO.
4.3.4 Cross-linking of PHO
Chain extension of PHO through chemical cross-linking was employed, in an attempt to increase
the viscosity of the dispersed phase to attain a more favourable viscosity ratio. Cross-linking of
MCL PHAs can be achieved using peroxides, radiation, or sulfur cures. Gagnon et al. [90] cross-
linked saturated and unsaturated MCL PHAs using four different types of peroxide with and
without coagents. They found that cross-linking decreased the crystallinity of the polymer.
Reduced tensile and tear strength was observed as a result of chain scission. Sulfur
vulcanization was also used by Gagnon et al. [17], whereas Dufresne et al. [91] cross-linked MCL
PHA by irradiation.
48
Figure 4.5 a) Un-notched impact strength and tensile strain b) tensile stress and Young’s
modulus of pristine PHB and its blends
0
2
4
6
8
10
12
14
0
2
4
6
8
10
12
14
16
0% 5% 10% 15% 20% 30%
Ten
sil
e S
train
(%
)
Un
no
tch
ed
Im
pact
(KJ/m
2)
PHO in PHB
Unnotched Impact
Tensile Strain
0
100
200
300
400
500
600
700
800
0
5
10
15
20
25
30
0% 5% 10% 15% 20% 30%
Yo
un
g's
Mo
du
lus
(M
Pa
)
Te
ns
ile
Str
es
s (
MP
a)
PHO in PHB
Tensile stress (MPa)
Young's Modulus(MPa)
a)
b)
49
In this work, given the sensitivity of the polymers to temperature, lauroyl peroxide was used,
because it decomposes at relatively low temperatures. A series of time sweeps were conducted
at various temperatures using the rheometer to generate a series of cure curves shown in
Figure 4.6. Based on these data, 155 °C was chosen for the cross-linking reaction, aiming for the
shortest possible reaction time.
Figures 4.3a-c show the effect of cross-linking on the linear viscoelastic properties of PHO.
Addition of peroxide increased significantly the complex viscosity and storage modulus of PHO,
whereas the loss tangent, tanδ, decreased below 1, revealing a transformation from viscoelastic
liquid to a viscoelastic solid.
Figure 4.6 Cure curves of PHO with 0.1 wt.% lauroyl peroxide as a function of temperature at a
frequency of 1 Hz.
10
100
1000
10000
0 500 1000 1500 2000
Sto
rag
e M
od
ulu
s (
Pa
)
Time (s)
135 ° C
145 ° C
155 ° C
50
A peroxide content of 0.06 wt.% was sufficient to achieve 67 wt.% gel content in the cross-
linked PHO, resulting in a tanδ value of about 1. Further increases in peroxide to 0.2 and 0.5
wt.% resulted in almost fully gelled material with gel contents of 87 and 97% respectively.
It should be noted that cross-linking resulted in a significant drop in the crystallinity of PHO, as
recorded from the 1st heating cycle, from 15 to 7% with 0.2 wt.% Lauroyl peroxide, whereas the
fully cross-linked PHO was completely amorphous.
The increases in viscosity upon cross-linking were accompanied by a substantial increase in
shear thinning behavior, and loss of the Newtonian plateau, as expected for cross-linked
polymers having high cross-link densities. Given the change in the viscosity-shear rate
dependence, cross-linked PHA can only match the viscosity of PHB in a very narrow
frequency/shear rate range. Based on the data of Figure 4.3, in order to match the viscosity of
PHO in the shear rate range of 10-100 s-1, which is relevant to compounding, 0.2-0.5 wt.% of
peroxide is needed.
The morphology of the blends containing 30 wt.% PHO cross-linked with different amounts of
lauroyl peroxide is compared to that of the unreacted blends in Figure 4.7. The corresponding
viscosity ratios, calculated at a representative shear rate of 50 s-1 from Figure 4.3a are shown in
the caption of Figure 4.7. The dispersed PHO domain size became progressively smaller upon
increasing the amount of cross-linking (Figure 4.7b and 4.7c). Significant improvement in
morphology was seen upon addition of 0.2 wt.% Lauroyl peroxide. It was impossible to assess
the domain size in the blends containing PHO cross-linked with 0.5 wt.% Lauroyl peroxide,
because the high gel content did not allow for etching and thus sufficient contrast (Figure 4.7d).
51
The improvement in morphology correlates well with the decrease in the viscosity ratio
achieved by using the cross-linked PHO dispersed phase.
Figure 4.7 SEM of blends of PHB/cross-linked PHO 70/30 blends a) uncross-linked (viscosity
ratio, λ=0.03) b) cross-linked with 0.06 wt.% of lauroyl peroxide (λ=0.15) c) cross-linked with 0.2
wt.% of lauroyl peroxide (λ=0.36) d) cross-linked with 0.5 wt.% of lauroyl peroxide (λ=3.73).
As shown in Table 4.2, there were significant improvements in the Young’s modulus and strain
at break of the blends containing cross-linked PHO, but the impact strength was unchanged.
This may be due to the high cross-link densities and gel content of the cross-linked PHO, which
alter its thermoplastic elastomer nature.
a) b)
)
a)
c)
)
b
a)
d)
)
b
a)
52
Table 4.2 Comparison of mechanical properties of PHB/PHO 70/30 blends containing uncross-
linked and cross-linked PHO
These results suggest that matching the viscosity of the blend components can result in finer
morphology and improvements in the mechanical properties. However high cross-link densities
and gel content of the cross-linked PHO, may present a limitation in terms of impact properties.
Therefore tight control of the morphology, while avoiding the formation of excessive gels, is
crucial to achieve desirable property improvements.
4.4 Conclusions
PHB/PHO blends had improved thermal stability, tensile strain at break and unnotched impact
strength compared to the unmodified PHB. The ability of PHO to act as an impact modifier for
PHB was limited by the viscosity mismatch between the two components, which resulted in a
coarse blend morphology. Chain extension of PHO by peroxide cross-linking improved the
PHB/PHO Tensile stress
(MPa)
Tensile strain
at break (%)
Young's modulus
(MPa)
Un-notched
impact (KJ/m2)
70/30 11.0 (±2.0) 9.6 (±0.5) 189 (±23) 14.2 (±0.4)
70/30 (0.2 wt.%
lauroyl peroxide ) 10 (±1.6) 14.4 (±0.3) 220 (±20) 13.3 (±1.4)
70/30 (0.5 wt.%
lauroyl peroxide ) 10.6 (±0.8) 14.5 (±0.4) 243 (±27) 14.2 (±0.5)
53
viscosity of PHO and led to better morphology and improved modulus and elongation at break
of the blends.
54
Chapter 5 Dramatic improvements in strain hardening and
crystallization kinetics of PLA by simple reactive modification in the
melt state*
5.1 Introduction
Poly(lactic acid)(PLA) is a bioderived, biodegradable thermoplastic polyester [92], which can be
processed using conventional thermoplastics processing equipment, including injection
molding, blow molding, film casting and blowing [5]. However it has a very narrow processing
window, because of the lack of melt strength and its slow crystallization rates. Additionally its
poor engineering properties, including impact strength and heat resistance have mainly
confined its applications to food packaging [38], as well as biomedical applications, such as drug
delivery, where biocompatibility and biodegradability are desired [93]. The properties of PLA
depend significantly upon its molecular weight and the stereochemical makeup of the
backbone, which is controlled by polymerization with D-lactide, L-lactide, or D,L-lactide, to form
random or block stereocopolymers [37]. Minimizing the amount of D-lactide is required to
obtain PLA with higher crystallinity; however most commercial grades have low crystallinities
and low crystallization rates, unless nucleating agents are used. The rheological properties of
PLA depend on the molecular weight and molecular weight distributions, presence of
branching, as well as its stereochemical makeup [94-98]. Strain hardening has been reported in
melts containing a high molecular weight tail [94], and in amorphous PLA containing mixtures
of the D and L isomers [96] at low temperatures, but otherwise it is generally accepted that
commercially available linear PLA lacks the level of strain hardening, and therefore melt
*A version of this chapter is accepted. Nerkar, M., Ramsay, J.A., Ramsay, B.A., Kontopoulou, M. Macromolecular Materials and Engineering, accepted May2014.
55
strength, needed for normal processing operations. This restricts its processability in operations
involving high stretch rates, such as film blowing, thermoforming, foaming etc. Given these
shortcomings, approaches have been proposed to achieve chain extension and/or branching in
PLA, with branching generally considered more beneficial [99]. Compared to the various
synthetic routes that exist to synthesize branched PLA, methods that employ reactive
modifications in the melt state are generally considered to be more convenient and industrially
relevant. Various modification approaches to improve processability have been summarized by
Pilla et al. [100] and Yu et al. [101]. These include chain extension in the presence of glycidol
[46] and long chain branching via functional group reactions of pyromellitic dianhydride and
triglycidyl isocyanurate [99]. Furthermore chain extenders, such as tris (nonylphenyl)
phosphate, polycarbodiimide and multi-functional epoxy compounds have been used to
counteract degradation in polyesters, such as PLA and to achieve chain extension
[47,51,102,103].
Reactive extrusion of PLA using organic peroxides has been undertaken to increase the
molecular weight, viscosity and melt strength with limited success, as the resulting branching is
often counter-balanced by severe chain scission [52-54]. Radiation induced cross-linking in the
presence of multi-functional coagents has been suggested as an alternative, but generally
resulted in physical property reduction [104].
Peroxide-initiated reactive extrusion in the melt state, assisted by coagents is frequently
employed as a means to introduce long-chain branching in linear polymers, such as
polypropylene[105,106]. However there are only two reports, employing this approach in PLA.
Yang et al. [107] used triallyl isocyanurate as a cross-linking agent together with dicumyl
56
peroxide (DCP), to obtain compounds with different levels of cross-linking. More recently, You
et al. [108] reported that PLA prepared through reaction with DCP and pentaerythritol
triacrylate (PETA) coagent had enhanced viscoelastic properties, which was attributed to
branching. The resulting product had faster crystallization rates under isothermal conditions.
However in these publications there was no mention about the properties of the resulting
materials under uniaxial extension and the non-isothermal crystallization behavior of the
polymers, which is relevant to processing, was not reported.
This chapter reports substantial improvements in the melt strength and non-isothermal
crystallization kinetics, upon employing a simple chemical modification method in the melt state
using solvent-free, peroxide-initiated grafting of a multi-functional coagent (triallyl trimesate,
TAM). To the best of our knowledge this is the first time that simultaneous improvements in all
these properties upon reactive modification are reported for PLA. These attributes are expected
to enable use of these materials in operations such as foaming, injection moulding and film
processing.
5.2 Experimental
PLA (grade 3251D, MFI 35 g/10 min at 190 °C/ 2.16 kg) was obtained from Natureworks®. TAM
(98%, Monomer Polymer Inc.), dicumyl peroxide (DCP, 98%, Sigma-Aldrich), acetone (Sigma-
Aldrich) and tetrahydrofuran (THF, Sigma-Aldrich) were used as received. Joncryl® ADR 4368 a
multi-functional epoxide styrene-acrylic oligomeric chain extender, containing glycidyl
methacrylate (GMA) functions was supplied by BASF. It has a functionality of 9 with epoxy
equivalent weight 285 g/mol, and molecular weight 6,800 g/mol [51].
57
PLA was dried in a vacuum oven at 100 °C for 3 h to remove moisture. Peroxide-degraded PLA
(PLA/DCP) was prepared by coating ground PLA powder (15 g) with an acetone solution
containing DCP (0.045 g) and allowing the solvent to evaporate. The resulting mixture,
containing 0.3 wt.% DCP, was charged to a DSM micro-compounder, equipped with twin-co-
rotating screws, at 180 °C at 100 rpm for 6 min. Coagent-modified PLA was prepared as
described for PLA/DCP from a mixture of PLA (14.8 g), DCP (0.045 g) and TAM (0.15 g), yielding
PLA/TAM containing 0.3 wt.% DCP and 1 wt.% TAM. A compound containing 1.2 wt.% of GMA,
which was the amount required to yield similar zero shear viscosity as the PLA/TAM was also
prepared and used for comparison. Neat PLA was processed under the same conditions
outlined above, to provide a suitable basis for comparison. After compounding, the strands
were quenched in cold water before chopping into pellets for further characterization.
Samples were prepared for size exclusion chromatography (SEC) analysis by dissolving 10 mg of
polymer in 1 mL of distilled THF overnight to ensure complete dissolution, and filtered through
a 0.2 µm nylon filter. Polymer molecular weight distributions (MWD) were determined with
respect to polystyrene standards using a Viscotek 270max separation module with triple
detection by differential refractive index (DRI), viscosity (IV) and light scattering (low angle LALS
and right angle RALS), which was maintained at 40 °C and contained two porous PolyAnalytik
columns in series. Distilled THF was used as the eluent at a flow rate of 1 mL/min.
The linear viscoelastic properties were measured in the oscillatory mode using a stress
controlled rheometer (Visco Tech by Reologica). Frequency sweeps were conducted at 180 °C
using 20 mm parallel plates, under nitrogen purge. Samples were further characterized in
uniaxial extension using an SER-2 universal testing platform from Xpansion Instruments hosted
58
on an MCR-301 Anton Paar rheometer. Measurements were conducted at 180 °C at Hencky
strain rates ranging from 0.10 to 10 s-1. The linear viscoelastic (LVE) oscillatory measurements
obtained at 180 °C were used to calculate the LVE stress growth curve, η+, and to check the
consistency of the extensional measurements. The curve corresponding to 3η+ represents the
LVE envelope in uniaxial extension, according to Trouton’s law.
Differential scanning calorimetry (DSC) was conducted using a DSC Q 100 by TA Instruments.
Samples were scanned between 0 and 200 °C at a heating rate of 5 °C/min. After the first
heating, each sample was held isothermally at 200 °C for 3 min before cooling at rates between
2.5 and 20 °C/min, to determine the crystallization onset and peak temperatures according to
ASTM D3418. The % crystallinity of PLA, was estimated using Equation (5.1)
100100
H
HHX ccm
c (5.1)
where ΔHm is the apparent fusion enthalpy, ΔHcc is the exothermic enthalpy that is absorbed by
crystals formed during the heating scan and ΔH100 is the theoretical fusion enthalpy of a 100%
crystalline polymer, which is 93.6 J/g for PLA [109].
Isothermal studies involved heating the sample to 200 °C and holding it for 3 min, followed by
cooling at 50 °C/min to temperatures ranging from 135-155 °C, where they were held
isothermally until completion of crystallization. The analysis included evaluations of the relative
crystallinity as a function of time and standard Avrami kinetics [110].
59
5.3 Results and Discussion
5.3.1 Rheological characterization
The linear viscoelastic properties of neat PLA, PLA reacted with 0.3 wt.% DCP (PLA/DCP) and
with 0.3 wt.% DCP/ 1 wt.% TAM (PLA/TAM) are summarized in Figure 5.1. PLA/DCP had
essentially unaltered flow characteristics compared to the unmodified PLA (Figure 5.1a) and
linear architecture, as revealed by the Van Gurp-Palmen plots of Figure 5.1b, tend to the limit of
90°. This suggests that peroxide-induced degradation was compensated by chain extension,
without branching. This is corroborated by the minimal differences in the molar mass
distributions between these two compounds (Table 5.1).
On the contrary, PLA/TAM demonstrated a substantial increase in molar mass (Table 5. 1) and
melt viscosity (Figure 5.1a). It is well-known that solvent-free processing with multi-functional
coagents involves simultaneous chain scission and cross-linking, the balance of which controls
the molecular weight and branching distributions of the final product. In general, reaction with
coagents bearing multiple acrylate, allylic or styrenic groups results in bimodal molecular
weight and branched distribution, comprised of a linear chain population of relatively low
molecular weight, and a high molecular weight hyper-branched chain population, which can
progress above the gel point [56,111]. Furthermore it has been reported that systems
containing polypropylene reacted with coagent and peroxide can undergo a precipitation
polymerization, which results in the formation of a separate phase of highly cross-linked,
coagent-rich sub-micron sized particles [56,112]. The resulting products possess a creamy
appearance in the melt state, owing to the presence of these cross-linked nano-particles [56].
60
Figure 5.1 a) Complex viscosity as a function of frequency and b) phase degree as a function of
complex modulus at 180 °C.
In the present case, treatment with 0.3 wt.% DCP and 0.1 wt.% TAM produced a creamy (as
opposed to the transparent PLA and PLA/DCP), gel-free product with increased melt elasticity,
and shear thinning, as observed in Figure 5.1, consistent with the presence of branching [57].
We compared the properties of PLA/TAM to those of PLA chain extended using a multi-
100
1000
10000
0.1 1 10 100 1000
Co
mp
lex
Vis
co
sit
y (
Pa
s)
Frequency (rad/s)
40
50
60
70
80
90
100
100 1000 10000 100000
Ph
as
e A
ng
le ( )
Complex Modulus (Pa)
PLA
PLA/DCP
PLA/TAM
PLA/GMA
b)
)
b
a)
a)
)
b
a)
61
functional epoxide styrene-acrylic oligomeric chain extender, containing GMA functions (trade
name Joncryl® from BASF).
Table 5.1 Material characterization
Material Mwa)
[g/mol]
PDIb) TM
[°C]
TC
[°C]
TC,onset
[°C]
TCC
[°C]
Crystallinity
[%]
PLA 98,140 1.7 173 N/A N/A 109 24
PLA/DCP 90,600 1.8 170 N/A N/A 94 35
PLA/TAM 143,020 2.0 169 133 149c), 142d) N/A 52
PLA/GMA 115,000 1.7 168 105 124c), 123d) 95 34
a)Mw: Weight average molar mass; b)PDI: polydispersity index; c)at cooling rate of 2.5 °C, d)at cooling rate of 5 °C.
The chain extended (PLA/GMA) was prepared by reacting PLA with 1.2 wt.% Joncryl®, which
was the amount needed to match the zero shear viscosity of PLA/TAM. The epoxy functions
contained within multi-functional epoxies can react with the –OH and –COOH end groups of
PLA, resulting in random branching and/or cross-linking [51]. The low levels of Joncryl® used
herein, produced a gel-free product with higher molar mass than the parent PLA (Table 5.1)
increased viscosity and deviations from the terminal flow behavior (Figure 5.1).
The different shear thinning characteristics and shapes of the Van Gurp-Palmen plots of
PLA/GMA compared to PLA/TAM point to different branching levels and chain topologies.
PLA/TAM presumably contained small amounts of a hyper-branched population, which are
characteristic of this coagent modification, whereas reaction with multi-functional epoxies
62
produces random branching. In spite of the different mechanisms of chain extension/branching,
the ultimate strain hardening characteristics of PLA/TAM and PLA/GMA were very similar, as
shown in Figure 5.2. Pronounced strain hardening was present in both materials, providing
evidence of long chain branching. Note that the viscosities of the parent PLA and PLA/DCP were
below the threshold needed to support extensional viscosity experiments.
Figure 5.2 Tensile stress growth coefficient (ηE+) of TAM and GMA modified PLA as a function of
strain rate and time at Hencky strain rates of 0.1, 1 and 10 s-1 at 180 °C. Curves are shifted by an
arbitrary factor for the sake of clarity. Solid lines represent the LVE envelop (3E+) for each
sample.
63
5.3.2 Thermal properties
Detailed DSC data are presented in Table 5.1 and Figure 5.3. The glass transition temperature
(Tg) of PLA was 62 °C and remained unchanged in all modified materials. PLA had a cold
crystallization peak, Tcc, at 109 °C, a melting peak, TM, at 173 °C (Table 5.1 and Figure 5.3a).
PLA/DCP had reduced cold crystallization temperature and increased crystallinity. These findings
are commonly associated to PLA degradation [51,108]. Neither of these two materials
crystallized during the cooling cycle.
Figure 5.3 DSC a) 2nd heating scan at rate of 5 °C/min b) cooling scan at the rate of 5 °C/min
70 90 110 130 150 170 190
Hea
t F
low
A
.U.
Temperature ( C)
PLA
PLA/GMA
PLA/TAM
70 90 110 130 150
He
at
Flo
w A
.U.
Temperature ( C)
PLA/DCP
PLA
PLA/GMA
PLA/TAM
PLA/DCP
a)
b)
64
On the contrary, the branched PLAs showed exothermal crystallization peaks. PLA/GMA had a
weak crystallization peak, TC, around 105 °C, suggesting a moderate effect of this modification
on the ability of the chains to crystallize. On the other hand, coagent modified PLA had a clear
and sharp crystallization peak at 133 °C (Figure 5.3b). This was accompanied by the
disappearance of the cold crystallization peak, and a significant increase in crystallinity by 117 %
with respect to neat PLA and 50 % with respect to PLA/DCP. Even though changes in the cold
crystallization of PLA upon modification with a PETA coagent have been mentioned previously
[108] this is the first time that the presence of an exothermic crystallization peak arising during
the cooling cycle, which is indicative of the capability of the material to crystallize upon cooling
during normal polymer processing operations, is reported.
The ability of our modified materials to crystallize was evident not only by the appearance of an
exothermic crystallization peak, but also by their isothermal and non-isothermal crystallization
kinetics (Figure 5.4). The PLA and PLA/DCP formulations did not crystallize and therefore are
not included in this comparison. Plots of the evolution of relative crystallinity as a function of
time revealed a crystallization half-time (t1/2) of 9.3 min at 135 °C for PLA/GMA, whereas the
half-time of PLA/TAM at this temperature was only 0.6 min. The results of the Avrami analysis
for temperatures ranging from 135-155 °C are presented in Table 5.2. The Avrami exponents
suggest similar crystal growth habit in all cases. Introduction of branching in polymers such as
polypropylene has been associated previously to changes in the crystallization kinetics [113].
The relative crystallinity as a function of time, recorded during non-isothermal crystallization
experiments is shown in Figure 5.4b. While cooling from the melt state, PLA/TAM started to
65
crystallize significantly earlier, with t1/2 of 2.6 and 1.8 min at cooling rates of 2.5 and 5°C/min
respectively, as compared to 8.4 and 3.9 min for PLA/GMA.
Figure 5.4 Relative degree of crystallinity as a function of time a) isothermal crystallization
experiments; (-) PLA/TAM at 135 °C, ()PLA/TAM at 140 °C, ()PLA/TAM at 150 °C, (◆)
PLA/GMA at 135 °C and (b) non-isothermal crystallization experiments; ()PLA/TAM at 2.5
°C/min, (◆)PLA/TAM at 5 °C/min, (o)PLA/TAM at 20 °C/min, ()PLA/GMA at 2.5 °C/min,
()PLA/GMA at 5 °C/min
0
25
50
75
100
0 2 4 6 8 10
Rela
tive
de
gre
e o
f c
rys
tall
init
y (
%)
Time (Min)
0
25
50
75
100
0 5 10 15 20
Re
lati
ve
de
gre
e o
f c
rys
tall
init
y (
%)
Time (Min)
a)
b)
66
Furthermore PLA/TAM had t1/2 of 0.96, 0.73 and 0.66 min at cooling rates of 10, 15 and 20 °C
respectively, whereas PLA/GMA did not crystallize at these conditions.
Table 5.2 Isothermal Avrami constants and crystallization half time for PLA/GMA and PLA/TAM
at various temperatures
Temperature [°C] n K [min-1] t1/2 [min]
PLA/GMA 135 2.9 0.0001 9.3
PLA/TAM 135 3.0 4.27 0.6
PLA/TAM 145 3.4 0.14 1.6
PLA/TAM 150 3.3 0.02 2.8
PLA/TAM 155 2.6 0.01 6.1
The findings reported above point to a nucleating effect, which occurred in spite of the absence
of a nucleating agent. As explained earlier, a nucleating effect attributed to the formation of a
separate phase of coagent-rich particles that forms upon reactive modification as a result of
TAM oligomerization [55], was reported recently in coagent modified polypropylene. We
suggest that a similar nucleation effect is responsible for the enhancements in crystallinity and
crystallization rates in the reactively modified PLA/TAM product.
5.4 Conclusions
PLA with long chain branching was produced by a simple radical mediated peroxide-initiated
grafting of TAM coagent in the melt state. The resulting product demonstrated strain hardening,
67
consistent with its long chain branched characteristics, and significantly enhanced crystallinity
and crystallization rates, suggesting that this is a promising approach to enhance the processing
characteristics and properties of PLA.
68
Chapter 6 Improvements in the extensional rheology, thermal
properties and morphology of poly(lactic acid)/ poly-3-
hydroxyoctanoate blends through reactive modification
6.1 Introduction
Among the key challenges associated with more wide-spread acceptance of biopolyesters, such
as poly(lactic acid) (PLA) and polyhydroxyalkanoates (PHA) in engineering applications are their
high production costs, brittle nature, slow crystallization rates and low melt strength, which
restrict their processability under common polymer processing operations, as well as their
hygroscopic nature, and susceptibility to degradation.
The advantages and drawbacks of PLA and PHAs as well as the current state-of-the-art of the
various modification methods that have been employed to overcome their limitations have
been reviewed recently [24,36,38,79,104,114]. Blending with ductile polymers and addition of
plasticizers are commonly used to improve the properties and processability of PLA [36,114],
and poly-3-hydroxybutyrate (PHB) [79,115]. Furthermore, to achieve fully bio-based
formulations, blends of biopolymers have been studied extensively [79,80,116].
PHAs and their copolymers have been used to enhance the toughness of PLA through solution
blending [117] or melt blending [117-119]. As already shown in Chapter 4, medium-chain-
length (MCL) PHAs can serve as impact modifiers for PHB, due to their low crystallinity and
elastomeric character [84,100,101].
To address the problem of lack of melt strength and suitable rheological properties of
biopolyesters, reactive modifications in the melt state have been employed to achieve chain
extension. Various modification approaches of PLA have been summarized by Pilla et al. [100],
69
and Yu et al. [101]. These include chain extension of PLA in the presence of glycidol [46] and
introduction of long chain branching via functional group reactions of pyromellitic dianhydride
and triglycidyl isocyanurate [99]. Furthermore chain extenders, such as tris (nonylphenyl)
phosphite, polycarbodiimide and epoxy-functionalized oligomeric acrylic copolymer (trade
name Joncryl® from BASF) have been considered to counteract degradation in PLA and
introduce chain extension [47,102,103]. In-situ cross-linked hyperbranched polymers have been
used to improve the toughness of PLA [120,121]. Reactive extrusion of PLA using organic
peroxides and coagents has also been undertaken [52-54,104,107,108]. On the other hand, as
discussed in Chapter 4, cross-linking of MCL PHAs can be achieved using peroxides, radiation, or
sulfur cures [17,90,91].
Reactive blending of biopolyesters, such as PHB and polyhydroxybutyrate-co-valerate (PHBV)
with polybutylene succinate (PBS) [31], and PLA with polycaprolactone (PCL) [122], PLA with
PBS [123], PLA with Poly(butylene adipate-co-terephthalate) (PBAT) [124] and PHB [125] using
peroxides has been employed to produce in-situ compatibilized blends having improved
properties. Another proposed chemical modification involved blending of PLA with MCL-PHAs
using diisocyanate chain extenders, which are highly toxic compounds [4]. Epoxy functionalities
have been introduced in MCL PHA to react with the hydroxyl groups of PLA, thus increasing the
interfacial interaction and improving the blend morphology and compatibility [117]. This
approach was also tested in the present thesis (Appendix A) with limited success.
In chapter 5, we presented a simple reactive modification approach, utilizing solvent-free,
peroxide-initiated grafting of a multi-functional coagent, to introduce branching and achieve
substantial improvements in the strain hardening characteristics of PLA. This approach resulted
70
in faster crystallization kinetics, both under isothermal and under non-isothermal conditions.
The present chapter investigates chain extension of PHO and PLA/PHO blends using the same
approach. The properties of the reactively modified blends are compared to those of
unmodified blends.
6.2 Experimental
6.2.1 Materials
Polyhydroxyoctanoate (PHO) containing ~98 mol% of 3-hydroxyoctanoate and ~2 mol% of 3-
hydroxyhexanoate was produced from glucose and octanoic acid, using bacterial fermentation,
as described by Xuan et al. [70,72]. The weight average molecular weight of the PHO,
determined by triple-detector size exclusion chromatography (SEC), was 172,000 Da with a
dispersity of 1.75 [126]. PLA (grade 3251D, MFI 35 g/10 min at 190 °C/ 2.16 kg) was obtained
from Natureworks®. Triallyl trimesate (TAM, 98%, Monomer Polymer Inc.), dicumyl peroxide
(DCP, 98%, Sigma-Aldrich), acetone (Sigma-Aldrich), and tetrahydrofuran (THF, Sigma-Aldrich)
were used as received.
6.2.2 Compounding
PLA and PHO were dried in a vacuum oven at 100 °C and at room temperature respectively, to
remove moisture. PLA/PHO blends containing 0-20 wt.% PHO were compounded in a DSM
microcompounder at 180 °C for 3 min at a screw speed of 100 rpm. The compounder was
operated under nitrogen blanket to limit polymer degradation. After compounding, the strands
were quenched in cold water before chopping into pellets. The neat materials were
compounded under the same conditions for comparison.
71
Peroxide-degraded PHO and PLA/PHO blends were prepared by coating ground PHO and PLA
powders with an acetone solution containing DCP and allowing the solvent to evaporate. The
resulting mixtures of PHO were charged to a DSM micro-compounder, equipped with twin-co-
rotating screws, at 180 °C at 100 rpm for 6 min. Coagent-modified PHO and PLA reacted with
DCP and TAM were prepared as described above from PHO or PLA, yielding various
compositions containing 0.2-0.5 wt.% DCP and amounts of TAM ranging from 0.5-2 wt.%.
Similarly coagent-modified PLA/PHO blends containing DCP and TAM were prepared. The
various compounds are designated using the name of the polymer, followed by the amounts of
DCP and TAM (i.e. PHO/0.3/1 denotes PHO reacted with 0.3 wt.% DCP and 1 wt.% TAM).
The gel content of coagent modified PHO and PLA/PHO blend was measured by dissolving the
material in chloroform for 7 h. The polymer was sealed in stainless steel wire (120 mesh)
according to ASTM D 2765. The material was left to stand for 1 h and subsequently dried
overnight in a vacuum oven at room temperature. The % gel content was calculated using
equation (6.1).
100sample of weight Initial
sample of weight Finalcontent Gel (6.1)
6.2.3 Characterization
6.2.3.1 Rheology
Compression molded discs, 25 mm diameter and 2 mm thick, were prepared using a Carver
press. The linear viscoelastic properties were measured in the oscillatory mode using a stress
controlled rheometer (Visco Tech from Reologica). Frequency sweeps were conducted at 180 °C
using 20 mm parallel plates.
72
Reactively modified PLA/PHO blends were further characterized in simple uniaxial extension
using an SER-2 universal testing platform from Xpansion Instruments hosted on the MCR-301
Anton Paar rheometer. Measurements were conducted at 180 °C at Hencky strain rates ranging
between 0.10 and 10 s-1. Specimens were prepared by compression molding the polymer
samples between polyester films to a gauge of about 0.75 mm, using a hydraulic press.
Individual polymer specimens were then cut to a width of 10 mm. Linear viscoelastic (LVE)
oscillatory measurements obtained at 180 °C were used to calculate the LVE stress growth
curve and check the consistency of the extensional measurements.
6.2.3.2 Differential scanning calorimetry (DSC)
DSC experiments were performed using a Q100 DSC from TA Instruments, under dry nitrogen.
Since MCL PHAs crystallize slowly, the samples were preconditioned to eliminate their thermal
history as follows: the polymer was heated at 100 °C for 10 min in a convection oven, and then
kept at room temperature for two weeks before characterization. Samples weighing 10-12 mg
were sealed in aluminum hermetic pans, equilibrated at -70 °C and kept isothermally for 5 min.
Afterwards they were heated to 200 °C at a rate of 5 °C/min and held isothermally for 3 min
before cooling to -70 °C at a rate of 5 °C/min. The samples were finally reheated to 200 °C at a
rate of 5 °C/min. The % crystallinity of the polymers, Xc, was estimated using equation (6.2).
100100
H
HHX ccm
c (6.2)
where, ΔHm is the enthalpy of fusion, ΔHcc is the exothermic enthalpy (cold crystallization)
recorded during DSC heating cycle and ΔH100 is the theoretical fusion enthalpy of a 100%
crystalline polymer. The heat of fusion for 100% crystalline PLA is 93.6 J/g [109].
73
6.2.3.3 Heat deflection temperature (HDT)
Specimens (127 mm X 13 mm X 3 mm) were prepared by compression molding using a Carver
press under 5000 N force, at 200 °C with a residence time of 3 min, then quenched in cold
water. Specimens were lowered in a silicon oil bath and the temperature was raised from 23 oC
at a heating rate of 120 oC/h. until 0.25 mm deflection occurred under a load of 1.82 MPa, in
accordance with ASTM D 648. At least three specimens were tested and the average value was
reported.
6.2.3.4 Mechanical properties
Specimens for mechanical property characterization were prepared by compression molding
using a Carver press under 5000 N force, at 200 °C and a residence time of 3 min, then
quenched in cold water. All specimens were conditioned at room temperature for 48 h after
compression molding, prior to mechanical testing. Tensile tests were conducted in accordance
with ASTM D638 using standard type V test specimens, with an Instron 3369 Universal tester, at
a cross head speed (CHS) of 5 mm/min. The average of five measurements is reported. Un-
notched Izod impact tests were conducted in accordance with ISO 180 using standard
specimens on a SATEC Instron machine and the average of five specimens are reported.
6.2.3.5 Scanning electron microscopy
Blend morphologies were observed using a JEOL JSM-840 scanning electron microscope.
Samples were first hot-pressed at 200 oC for 3 min, then immersed in liquid nitrogen for 3 min
before brittle fracture. The MCL PHA phase was etched in acetone overnight at room
temperature. The coagent modified blend samples were examined without etching as MCL PHA
was partially cross-linked and could not be dissolved.
74
6.2.3.6 Hot stage microscopy
Isothermal crystallization experiments were performed using a Linkam CSS 450 hot stage
mounted on an Olympus BX51 optical microscope. The sample was first heated to 200 °C at a
rate of 30 °C /min and held for 10 min to eliminate the heat history. The melt was then cooled
to 135 °C at 30 °C/min. The crystallization process was recorded isothermally at 135 °C using a
Sony ExwaveHAD 3 CCD digital recorder.
6.3 Results and Discussion
6.3.1 Blends of PLA with PHO
PLA and PHO form an immiscible blend system, having droplet-matrix morphology at
compositions up to 30 wt.% PHO, as shown in Figure 6.1. These blends have very coarse
morphology, with the average dispersed domain size changing from 1.5 ( 0.2) µm for 95/05
PLA/PHO blend to 5 ( 0.6), 7 ( 2) and 7.6 ( 2) µm for 90/10, 85/15 and 80/20 blend
respectively. This is consistent with the findings reported in Chapter 4 on PHB/PHO blends, and
is attributed to the significant viscosity mismatch between the blend components. PLA and PHO
had an almost Newtonian behavior, with viscosities of 660 Pa.s and 13 Pa.s for PLA and PHO
respectively, at the compounding temperature of 180 °C, resulting in a viscosity ratio (defined
as the ratio of the viscosity of the dispersed phase over the viscosity of the matrix) of 0.02.
Generally viscosity ratios as close as possible to 1 are required to achieve optimum blend
morphology, whereas viscosity ratios much higher or lower than 1 result in coarse
morphologies and tendency toward coalescence during melt compounding.
75
Table 6.1 summarizes the mechanical properties of the blends. Significant improvement was
observed in the elongation and impact properties, accompanied by a decrease in Young’s
modulus and tensile strength. The decline in these properties was associated with the decrease
in crystallinity of the blend, which is expected because of the addition of an amorphous minor
phase.
Figure 6.1 Scanning electron microscope images of PLA blend containing a) 5 wt.%, b) 10 wt.%,
c) 15 wt.% and d) 20 wt.% of PHO.
From Table 6.1 it is obvious that the optimum levels of impact strength were obtained at
compositions between 10 and 15 wt.% PHO, whereas the values decreased when higher
76
amounts were added. This is attributed to the morphology of the blend, which became much
coarser at compositions above 10 wt.% PHO, as shown in Figure 6.1.
Table 6.1 Mechanical properties of PLA and PLA/PHO blends
As mentioned in the previous chapters, low melt viscosities, viscosity mismatch and absence of
melt strength are factors that make processing of biopolyesters and their blends difficult,
resulting in a need for chain extension. In Chapter 4, we proposed peroxide-mediated cross-
linking of the PHO dispersed phase to achieve higher viscosity, and therefore to improve the
morphology of the blend. However, this approach produced high gel contents and limited
property improvements in blends with PHB. A similar attempt at blending peroxide cross-linked
PHO with PLA, shown in Appendix A, also resulted in limited success.
In the present chapter, a reactive modification procedure was implemented using peroxide and
coagent, similar to the approach described in Chapter 5, to achieve branching and therefore
PHO
( wt.%)
Tensile stress
(MPa)
Young's
Modulus
(MPa)
Elongation
at break (%)
Unnotched
Impact
(KJ/m2)
Crystallinity
(%)
0 74 (±3) 670 (±74) 14 (±1) 32 (±5) 24
5 56 (±4) 582 (±68) 24 (±10) 63 (±6) 17
10 50 (±4) 696 (±29) 35 (±15) 65 (±5) 16
15 45 (±3) 571 (±22) 47 (±10) 53 (±9) 16
20 34 (±3) 442 (±34) 24 (±3) 40 (±5) 16
77
chain extension, while avoiding excessively high gel contents. The following sections describe
the effects of coagent modification on PHO, PLA and PLA/PHO blends.
6.3.2 Reactive modification of PHO
PHO was reacted with various amounts of TAM and DCP, aiming in general at choosing
formulations that would use the least amounts of reagents possible, while still achieving
acceptable improvements in viscosity and elasticity, without excessive gel contents.
As shown in Figure 6.2, the flow characteristics of PHO remained unaltered when reacted with
0.3 wt.% DCP (PHO/0.3). The absence of changes in the presence of DCP is opposite to what
was observed in Chapter 4, when lauroyl peroxide was used. This suggests that in the presence
of DCP chain extension/cross-linking in the presence of peroxide is counteracted by significant
chain scission, possibly because of the higher compounding temperatures needed for DCP
compared to lauroyl peroxide.
When the multifunctional coagent, TAM, was added to the formulation, significant changes
were noted in the rheological properties of PHO. Two distinct groupings are noted in Figure
6.2a. When TAM amounts up to 0.75 wt.% were added (sample PHO/0.3/0.75), the increase in
complex viscosity was relatively small. On the contrary, addition of 1 wt.% TAM resulted in a
pronounced increase in viscosity, shear thinning behavior, and elasticity. Above this level of
TAM, the properties showed a tendency to plateau.
78
Figure 6.2 Effect of TAM content on the rheological properties of PHO with DCP content
remaining constant a) complex viscosity b) storage modulus and c) tan δ
1 10 100
Co
mp
lex
Vis
co
sit
y (
Pa
s)
104
103
102
101
100
1 10 100
Sto
rag
e M
od
ulu
s (P
a)
104
102
103
101
10-2
10-1
100
1 10 100
tan
δ
Frequency (rad/s)
PHO PHO/0.3PHO/0.3/0.5 PHO/0.3/0.75PHO/0.3/1 PHO/0.3/2
103
102
100
10-1
101
b)
a)
c)
79
Reactive modification was accompanied by changes in the appearance of the extrudate. The
PHO formulations containing low amounts of TAM were sticky and did not form strands when
extruded. On the contrary, addition of 1 wt.% TAM resulted in the formation of strands that
were not sticky and were easy to handle for further processing and characterization as shown in
Figure 6.3.
Figure 6.3 a) unmodified PHO after extrusion b) PHO/0.3/1 after extrusion
In an effort to achieve an optimum formulation, the amount of DCP was varied, while keeping
the amount of TAM constant at 1 wt.% (Figure 6.4). From this figure, it is obvious that amounts
of DCP above 0.3 wt.% were necessary to obtain significant improvements in complex viscosity
and elasticity. However the material reacted with 0.5 wt.% DCP was highly cross-linked, with a
gel content of 42 %. Based on the above results, the formulation containing 0.3 wt.% of DCP
with 1 wt.% of TAM showed the best improvement in viscosity, while maintaining a moderate
gel content of 23%.
a) b)
80
Figure 6.4 Effect of DCP amount on a) complex viscosity b) storage modulus and c) tan δ of
coagent modified PHO (PHO 0.3/1 and PHO 0.5/1 yielded 23 and 42 % gel respectively)
1 10 100
Co
mp
lex V
isco
sit
y (
Pa s
)
Frequency (rad/s)
PHO/0.2/1
PHO/0.3/1
PHO/0.5/1
102
1 10 100
Sto
rag
e M
od
ulu
s (P
a)
Frequency (rad/s)
10-2
105
104
103
102
101
100
10-1
1 10 100
tan
δ
Frequency (rad/s)
103
105
104
103
101
100
102
101
100
10-1
10-2
a)
b)
c)
81
In spite of the dramatic changes in the rheological properties, the thermal properties of the
reactively modified PHO, including the glass transition temperature (Tg) remained unaltered,
whereas the material remained highly amorphous, with no obvious melting transition.
6.3.3 Reactive modification of PLA
The reactive modification of the PLA matrix to provide branched PLA has been described in
detail in Chapter 5. Figure 6.5 shows the rheological properties of PLA reacted with two
different amounts of coagent and comparison with PHO/0.3/1. The formulation reacted with 2
wt.% TAM had very high viscosity, and appeared fully crosslinked and hard to process. On the
contrary, PLA/0.3/1 had negligible gel content, therefore this composition was deemed suitable
for the blend compositions described in Section 6.3.3. below.
Figure 6.5 Effect of coagent modification on the complex viscosity of PLA and PHO
1 10 100
Co
mp
lex V
isco
sit
y
(Pa s
)
Frequency (rad/s)
PLA PLA/0.3/1
PLA/0.3/2 PHO/0.3/1
102
102
104
103
101
104
102102
102
102
104
103
101
104
101
82
The extensional properties of the PLA/0.3/1 are summarized in Figure 6.6. As explained in
Chapter 5, reactive modification resulted in substantial strain hardening, which provides solid
evidence of a branched structure.
Figure 6.6 Tensile stress growth coefficient (ηE+) of PLA/0.3/1 and (PLA/PHO)/0.3/1 as a
function of strain rate and time at Hencky strain rates of 0.1, 1 and 10 s-1 at 180 °C. Curves are
shifted by an arbitrary factor for the sake of clarity. Dotted lines represent the LVE envelop for
each sample.
In addition to the enhancements in viscosity and strain hardening, the coagent modified PLA
had substantially different thermal properties compared to the neat PLA, as shown in Table 6.2,
including the appearance of a sharp crystallization peak in the DSC exotherm (Figure 6.7),
disappearance of the cold crystallization peak and increase in crystallinity.
0.01 0.1 1 10 100
ȠE
+(P
a s
)
t (s)
0.1 s-11 s-110 s-1
PLA/PHO/0.3/1
105
106
103
104
PLA/0.3/1
10x
102
83
Table 6.2 Thermal properties of neat, DCP and coagent modified PHO, PLA and PHO/PLA blend
TM (°C) TC (°C) TCC (°C) Crystallinity (%)
PLA 173 NA 109 24
PLA/0.3 170 NA 94 35
PLA/0.3/1 169 133 NA 52
PLA/PHO 170 NA 105 16
(PLA/PHO) /0.3 170 NA 104 17
(PLA/PHO) /0.3/1 169 138 NA 56
As reported in Chapter 5, the crystallization kinetics of PLA appeared significantly enhanced,
both in isothermal, and non-isothermal experiments. This finding was attributed to the
formation of a separate phase of coagent-rich particles that forms upon reactive modification,
which results in a nucleating effect [55].
84
Figure 6.7 DSC (a) cooling exotherm (b) heating endotherm of coagent-modified PLA and
PLA/PHO blends
80 100 120 140 160 180
He
at
Flo
w A
. U
.
Temperature ( C)
PLA/PHO
PLA/PHO/0.3
PLA/PHO/0.3/1
PLA/0.3/1
80 100 120 140 160 180
He
at
Flo
w A
.U.
a)
b)
85
Further evidence of the altered crystalline structure of these materials upon coagent
modification is provided by the optical microscope images in Figure 6.8, which depict PLA and
PLA/0.3/1 samples crystallized under isothermal conditions at 135 °C. The unmodified PLA did
not show evidence of crystal structure formation when cooled for 5 min, whereas the coagent-
modified sample is characterized by the formation of a dense spherulitic structure.
Figure 6.8 Hot stage microscopy of a) PLA, b) PLA/0.3/1 at 135 °C
The profound changes in the thermal and rheological properties of PLA are expected to have a
major impact on the properties of reactively compounded blends of PLA with PHO, when PLA is
the matrix phase. These blends are presented in section 6.3.4 – 6.3.7 below.
6.3.4 Reactive compounding of PLA with PHO
Based on the analysis presented in section 6.3.1, blends of PLA/PHO containing 10 wt.% PHO,
were used as this was identified as the optimum composition in terms of mechanical
properties. An additional reason for keeping the PHO content as low as possible is its high cost.
For simplicity these blends will be noted below as PLA/PHO.
a) b)
86
6.3.5 Thermal and rheological properties
Reaction of PLA/PHO blends with various amounts of peroxide and coagent resulted in similar
trends as noted in the sections above, i.e. increase in the complex viscosity, loss of Newtonian
plateau, increased shear thinning and elasticity of the blends. The values of all viscoelastic
properties were in-line with those seen for the PLA matrix at equivalent contents (Figures 6.5
and 6.9), with what appears to be limited influence of the minor PHO dispersed phase.
Similarly, the extensional stress growth data showed strain hardening behavior, following the
same trends as those seen for the branched PLA matrix (Figure 6.6).
In agreement with the findings reported for the reactively modified PLA, the modified PLA/PHO
showed a clear and sharp crystallization peak at 138 °C (Figure 6.7a). This was accompanied by
the disappearance of the cold crystallization peak, and significantly higher crystallinities
compared to the unreacted materials (Figure 6.7b). The thermal properties of all the reactively
modified compounds and their unreacted counterparts are summarized in Table 6.2.
6.3.6 Blend morphology
Compounding of PLA and PHO in the presence of DCP alone, resulted in a significant reduction
in the PHO domain size, as seen in Figure 6.10. The size of the PHO dispersed domains
decreased from 5 ( 0.6) µm to 1 ( 0.2) and 1.2 ( 0.2) µm when 0.5 and 1 wt. % of DCP was
added. Since, based on the rheological evaluations presented above, DCP alone did not
influence the rheological properties of the constituents of the blend, the most likely
explanation for this observation is a compatibilizing effect in the presence of DCP. This could be
attributed to the formation of small amounts of copolymer, by cross-termination of free radical
chains of PLA and PHO during the compounding procedure. In-situ compatibilization by reactive
87
blending of biopolyesters through reaction with peroxides to produce blends with finer
morphologies has been reported previously [17,123-125].
Figure 6.9 Effect of DCP and TAM on a) complex viscosity b) storage modulus and c) tan δ of
PLA/PHO blends
1 10 100
Co
mp
lex
Vis
co
sit
y (
Pa
s)
Frequency (rad/s)
104
103
102
1 10 100
Sto
rag
e M
od
ulu
s (P
a)
Frequency (rad/s)
106
105
104
102
103
101
1 10 100
tan
δ
Frequency (rad/s)
PLA/PHO PLA/PHO/0.3
PLA/PHO/0.3/0.3 PLA/PHO/0.3/1
PLA/PHO/0.5/1
102
101
100
10-1
b)
a)
c)
88
Ma et al. [124,127] and Dong et al. [125] attributed the compatibilization effect to the
combination of macroradicals that form in the presence of DCP via hydrogen abstraction. The
macroradicals may further recombine to form complex products, including copolymer at the
interface, resulting in a compatibilization effect.
Addition of TAM to the formulation resulted in further refinement in the morphology, which is
may be attributed to the improved viscosity of the PHO. The PHO domain size reduced from 1.5
( 0.2) to 0.55 ( 0.08) µm and from 5 ( 0.6) to 1 ( 0.16) µm for 95/5 and 90/10 PLA/PHO
blends respectively. In case of 80/20 the blend morphology changed from droplet-matrix to co-
continuous. Figure 6.11 shows the altered morphologies obtained in the presence of TAM, in
blends containing various amounts of PHO. In addition to the finer morphology noted
previously, these blends exhibited a different, co-continuous morphology at a PHO content of
20 wt.%. Such changes in morphology are common in thermoplastic vulcanizates, consisting of
PP and EPDM [128] but have never been noted for these biopolyester blends.
Figure 6.10 Scanning electron microscopy of PLA/PHO (90/10) blend a) unmodified b)
(PLA/PHO)/0.5 c) (PLA/PHO)/1
20mm 20mm20mm
a) b) c)
89
Figure 6.11 Effect of coagent modification on morphology of PLA/PHO blends a) 95/05 b) 90/10
c) 80/20 (wt./wt.%) (samples reacted with coagent were not etched); Top row without coagent;
bottom row with coagent
6.3.7 Mechanical properties
The reactively modified PLA/PHO blend had better tensile strain and unnotched impact
strength compared to pristine PLA, however, its properties were not as good as the unmodified
blend, in spite of the improved morphology (Table 6.3). The gel content of coagent modified
PLA/PHO blend was 20 %, revealing the presence of cross-linked chains in the modified blend.
These gels might have compromised the ductility of the blend, thus explaining the drop in
ductility and impact strength compared to the unmodified blend. Addition of PHO to PLA, as
well as reactive modification, did not affect its heat deflection temperature (HDT) (Table 6.3).
a) b) c)
90
Table 6.3 Mechanical properties and heat deflection temperature of neat and coagent modified
PLA, and PLA/PHO blends
These findings suggest that avoiding gels is crucial for the optimization of the properties of
these materials. Cross-linking/gelation is also expected to affect negatively the biodegradability
of these materials. Therefore optimization of the compositions to avoid the formation of gels
should be a high priority.
These results have shown that coagent modification improves significantly the processability of
the materials, by improving the melt strength and crystallization rates, while the mechanical
properties remain relatively unaffected when the specimens are prepared under the same
conditions. The differences noted in thermal properties and morphology however suggests that
processing conditions during the solidification stage may be tuned to further impart change in
the mechanical properties. This should be a topic of further investigation.
Tensile
stress
(MPa)
Tensile
strain
(%)
Unnotched
Impact
(KJ/m2)
Young's
Modulus
(MPa)
HDT
(°C)
PLA 74 (±3) 14 (±1) 32 (±5) 670 (±74) 55
PLA/0.3/1 77 (±1) 13 (±2) 35 (±2) 837 (±74) 56
PLA/PHO 50 (±1) 35 (±15) 65 (±5) 696 (±7) 55
(PLA/PHO)/0.3/1 49 (±3) 24 (±4) 55 (±3) 677 (±34) 54
91
6.4 Conclusions
Addition of PHO to PLA increased the impact strength and elongation at break of PLA, at the
expense of the Young’s modulus, while the HDT remained unaffected. The droplet-matrix
morphology of the blends was coarse, because of the very low viscosity of PHO, resulting in a
viscosity mismatch between the blend components.
The viscosity of PHO was successfully increased through partial cross-linking, using solvent-free
chemical modification using DCP and TAM coagent. In addition to the substantial increase in
viscosity and melt elasticity, the resulting product exhibited improved extrudate appearance.
Reactively modified PLA/PHO in the presence of DCP and TAM displayed the enhancements in
strain hardening and crystallization rates, previously observed for the matrix modified PLA.
Furthermore these blends had finer morphology, which was attributed to a compatibilizing
effect possibly arising from copolymer formation at the interface. Coagent modification further
resulted in changes in the morphology, and possible phase inversion in blends containing higher
PHO contents.
92
Chapter 7 Thesis overview
7.1 Thesis overview
Ongoing need for alternate options to conventional petroleum based polymers has resulted in
significant attention to various biopolyesters, including poly(lactic acid) (PLA) and poly-(3-
hydroxyalkanoates) (PHAs), which are bioderived and biodegradable. However, most of these
materials typically suffer from high production costs, brittleness, slow crystallization rates and
low melt strength, which restrict their processability under common polymer processing
operations. This thesis focused on the improvement of properties and processability of
biopolyesters through blending and reactive compounding. Specifically the potential of
elastomeric medium-chain-length (MCL) PHAs, as potential impact modifiers for PLA and brittle
poly-3-hydroxybutyrate (PHB) was assessed, using conventional melt compounding.
In spite of the relatively high molar masses (ranging from 18,200 to 172,000 g/mole), MCL PHAs
melts have low viscosity, presumably due to their helical conformation. Melt blending of these
materials with PLA and PHB resulted in coarse blend morphologies due to the large viscosity
mismatch, however, remarkable improvement was observed in ductility and impact strength of
PHB and PLA. As expected these improvements were accompanied by a decline in tensile stress
and Young’s modulus. Free-radical mediated cross-linking of polyhydroxyoctanoate (PHO) using
a peroxide resulted in improved blend morphology, because of the increased viscosity of PHO;
however the impact properties did not show further improvements, presumably because of the
high gel content. Furthermore chain extension using epoxidized PHO (ePHO) were explored.
This approach lowered the interfacial tension between PLA and PHO, because of the
93
interactions between the epoxy groups of ePHO with the carboxylic acid end group of PLA, thus
resulting in improved compatibility. However the morphology remained coarse, because of the
extremely low viscosity of ePHO.
The inability of the previously suggested approaches to produce substantial improvements in
the processability and properties of these polyesters led to the development of a simple
reactive compounding approach, involving reaction with a peroxide and coagent. This method
resulted in substantially improved crystallization rates, viscosity, elasticity and melt strength of
PLA and its blends with PHO, as well as improved the compatibility between the blend
components, resulting in a very fine morphology. This approach has resulted in blends having
simultaneously improved toughness, increased viscosity, strain hardening and crystallinity and
represents a significant technological advance in these materials.
These results suggest that simple free-radical mediated reactive compounding of biopolyesters
in the melt state can produce the rheological enhancements needed in processes such as film
blowing and casting, blow molding, thermoforming and foaming, as well as the enhanced
crystallization rates needed in injection molding, thus significantly broadening the applicability
of these polymers in conventional polymer processing operations. This should enable the
introduction of these compounds to new products/markets and may lead to the possible
identification of new applications for these biopolymers.
7.2 Conclusions
Absolute molecular weight (MW) distributions were determined for different MCL PHAs with
predominantly 3-hydroxyoctanoate (PHO), 3-hydroxynonanoate (PHN) or 3-
hydroxydodecanoate (PHDD) content via triple-detector size exclusion chromatography (SEC),
94
combined with analyses using various detectors, using tetrahydrofuran (THF) as the carrier
solvent. Unlike with the short-chain-length (SCL) PHB, the uncorrected polystyrene calibration
in THF provided a good estimate (within 10 %) of absolute MW values for the tested MCL PHAs,
irrespective of side chain length. Weight-average MW values ranged from 172,000 Da for PHO
to 18,200 for PHN with 30 mol% 3-hydroxyheptanoate, and dispersities of all samples were
close to two. Melt viscosity data suggested an entanglement molecular weight around 8 X 104
Da, significantly higher than most thermoplastics.
Blends of PHO with PHB, were prepared by melt compounding. Coarsening of the droplet-
matrix morphology of the blends was noted as the PHO content increased beyond 5 wt.%; this
was attributed to the significant viscosity mismatch between the components. Addition of PHO
improved the thermal stability of the blends, reduced their crystallinity and resulted in shifts in
their melting and crystallization temperatures. The blends had improved tensile strain at break.
The unnotched impact strength showed a threefold increase at 30 wt.% PHO content. Cross-
linking of PHO using a peroxide initiator increased its viscosity, thus improving the morphology
and mechanical properties of the blends.
Blends of PLA with PHO were also investigated. In spite of the viscosity mismatch between the
blend components, addition of PHO resulted in increased elongation at break and impact
strength of PLA accompanying decrease in tensile strength and modulus. These improvements
can be explained by a reduction in the crystallinity of the PLA, whereas the melting point and
glass transition temperature (Tg) remained unaffected. Further increase of the PHO content
resulted in a drop in properties, and attributed to the coarse morphology. The heat deflection
temperature (HDT) of PLA did not change upon addition of 10 wt.% PHO. Peroxide cross-linking
95
of PHO to increase its melt viscosity, and introduction of epoxy groups led to only moderate
improvements in blend properties.
In order to improve melt and crystallization properties, PLA was chemically modified by radical
mediated solvent-free, peroxide-initiated grafting of triallyl trimesate (TAM) coagent in the
melt state. When compared with the parent material and with PLA samples treated with
peroxide alone, coagent-modified materials demonstrated higher molar mass and improved
melt rheological properties, including substantial improvements in melt elasticity and strain
hardening under uniaxial extension. The properties of coagent modified PLA were compared to
those PLA modified by a multi-functional epoxide oligomeric chain extender (Joncryl®).
Although the rheological properties were comparable, the coagent-modified material
demonstrated significantly enhanced crystallinity and crystallization rates. The appearance of a
distinct crystallization exothermic peak and the disappearance of the cold crystallization
temperature point to a nucleation effect in the coagent modified PLA, which together with the
rheological enhancements can promote the processability of this material in conventional
thermoplastics operations.
Reactive compounding in the presence of the dicumyl peroxide (DCP) and TAM was also
evaluated in PHO and its blends with PLA. The viscosity and elasticity of PHO increased
substantially following reactive compounding indicative of partial cross-linking, while it retained
its amorphous nature. In addition to the substantial increase in viscosity and melt elasticity, the
resulting product exhibited improved extrudate appearance. Reactively modified PLA/PHO
blends in the presence of DCP and TAM displayed enhancement in strain hardening and
crystallization rates, previously observed for the matrix, modified PLA. Furthermore these
96
blends had finer morphology, which was attributed to a compatibilizing effect possibly arising
from copolymer formation at the interface. Coagent modification further resulted in changes in
the morphology, and possible phase inversion in blends containing higher PHO content, while
the mechanical properties and HDT remained relatively unaffected by this coagent
modification.
7.3 Significant contributions
This thesis has completed a very thorough and complete characterization of various MCL PHAs,
which are possible candidates to replace conventional elastomeric polymers. For the very first
time, the true molecular weight of MCL PHAs was determined. Based on the findings, molar
mass determinations based on polystyrene standards can be used as the molecular weight of
these materials.
MCL PHAs have very low viscosities and crystallize very slowly, requiring 8-12 h to solidify after
processing. This makes their handling and post-processing very difficult. The work in this thesis
developed a method to improve processing and handling of MCL PHA, through a simple
reactive modification using a peroxide and coagent.
It was shown that PHO, a bioderived and biodegradable polymer, can be used to impact modify
brittle biopolymers to offer a complete biosystem. Addition of PHO improved elongation and
impact properties of brittle PLA and PHB.
The most important contribution of this thesis is that it provided a solution to improve the melt
strength of PLA and its blends with PHO. This should extend its use in applications such as blow
molding, thermoforming, blown film and foaming that involve stretching. The improvements in
the total crystallinity and crystallization rates of PLA and its blends with PHO without the need
97
to add nucleating agents, are another significant finding of this thesis, which will likely have
broad implications in their potential to extend to new applications.
7.4 Recommendation for future work
i) Optimization of coagent content is needed to reduce gel content so that the
elastomeric properties of MCL PHA can be retained and its ability to impact modify
brittle polymers can fully be utilized.
ii) Most of the MCL PHAs, with the exception of poly(3-hydroxydodecanoate) (PHDD)
were highly amorphous. It may be worthwhile to pursue coagent modification of
PHDD, to investigate the potential to impart some improved crystallinity to this
material.
iii) The solid-state properties of coagent modified MCL PHAs should be investigated, as
these materials could be used as replacements for conventional thermoplastic
elastomers.
iv) The effect of coagent modification on pristine PHB and its blends with MCL PHA can
be investigated in detail to see whether the properties of these blends can be
enhanced.
v) Further detailed investigations of the improvements in crystallization in the
presence of coagent should be carried out, including detailed optical microscopy, X-
ray diffraction (XRD) analysis, as well as investigation of coagent-rich particle
formation in these compounds. The improvements in the morphology of PLA/MCL
PHA blends in the presence of peroxide and coagent should also be studied.
98
vi) The mechanical properties of coagent modified materials should be influenced by
the differences in their crystallinity. Detailed investigation of this, including different
cooling rates during sample preparation, as well as different molding procedures,
(injection and compression molding) should be carried out to exploit this effect.
99
References
1. Azapagic A, Emsley A, Hamerton I. Polymer – The environment and sustainable development.
Hamerton I, Wiley, 2003.
2. Gerald S. Degradable polymers: principles and applications. Netherlands: Kluwer academic
publishers, 2002.
3. Opportunities in chemistry by National research council. Washington DC: National academy
press, 1985.
4. Lee J, McCarthy S. Biodegradable poly(lactic acid) blends with chemically modified
polyhydroxyoctanoate through chain extension. J Polym Environ 2009;17(4):240-247.
5. Lim L, Auras R, Rubino M. Processing technologies for poly(lactic acid). Progress in Polymer
Science 2008;33(8):820-852.
6. Hanggi U. Requirements on bacterial polyesters as future substitute for conventional plastics
for consumer goods. FEMS Microbiol Rev 1995;16(2-3):213-220.
7. Lemoigne M. Ann.Inst.Pasteur,Paris. 1925,(39):144.
8. Steinbuchel A, Valentin H. Diversity of bacterial polyhydroxyalkanoic acids. FEMS Microbiol
Lett 1995;128(3):219-228.
9. Takagi Y, Yasuda R, Yamaoka M, Yamane T. Morphologies and mechanical properties of
polylactide blends with medium chain length poly(3-hydroxyalkanoate) and chemically
modified poly(3-hydroxyalkanoate). J Appl Polym Sci 2004;93(5):2363-2369.
10. Gassner F, Owen A. Physical properties of poly(beta-hydroxybutyrate) poly(epsilon-
caprolactone) blends. Polymer 1994;35(10):2233-2236.
100
11. Grassie N, Murray EJ, Holmes PA. The thermal degradation of poly(-(d)-β-hydroxybutyric
acid): Part 1—Identification and quantitative analysis of products. Polym Degrad Stab
1984;6(1):47-61.
12. Hong S, Gau T, Huang S. Enhancement of the crystallization and thermal stability of
polyhydroxybutyrate by polymeric additives. J Therm Anal Calorim 2011;103(3):967-975.
13. Huisman G, Deleeuw O, Eggink G, Witholt B. Synthesis of poly-3-hydroxyalkanoates is a
common feature of fluorescent pseudomonads. Appl Environ Microbiol 1989;55(8):1949-1954.
14. Jiang X. Process development for the production and separation of medium-chain-length
poly ( 3-hydroxyalkanoates) by Pseudomonas putida KT2440. Ph.D. Thesis, 2010.
15. Diard S, Carlier J, Ageron E, Grimont P, Langlois V, Guerin P, Bouvet O. Accumulation of
poly(3-hydroxybutyrate) from octanoate, in different Pseudomonas belonging to the rRNA
homology group I. Syst Appl Microbiol 2002;25(2):183-188.
16. Sun Z, Ramsay JA, Guay M, Ramsay BA. Fermentation process development for the
production of medium-chain-length poly-3-hyroxyalkanoates. Appl Microbiol Biotechnol
2007;75(3):475-485.
17. Gagnon K, Lenz R, Farris R, Fuller R. Crystallization behavior and its Influence on the
mechanical properties of a thermoplastic elastomer produced by Pseudomonasoleovorans.
Macromolecules 1992;25(14):3723-3728.
18. Utracki LA. Polymer blends handbook. Kluwer academic publishers, 2002.
19. Parulekar Y, Mohanty A. Biodegradable toughened polymers from renewable resources:
blends of polyhydroxybutyrate with epoxidized natural rubber and maleated polybutadiene.
Green Chem 2006;8(2):206-213.
101
20. Avella M, Martuscelli E. Poly-d-(-)(3-hydroxybutyrate) poly(ethylene oxide) blends - phase-
diagram, thermal and crystallization behavior. Polymer 1988;29(10):1731-1737.
21. Hori Y, Takahashi Y, Hongo H, Yamaguchi A, Hagiwara T. Takasago Int. 1996(EP 723 983).
22. Kumagai Y, Doi Y. Journal of Environmental Polymer Degradation 1993;1(2):81.
23. de Carvalho FP, Quental AC, Felisberti MI. Polyhydroxybutyrate/acrylonitrile-g-(ethylene-co-
propylene-co-diene)-g-s tyrene blends: Their morphology and thermal and mechanical
behavior. J Appl Polym Sci 2008;110(2):880-889.
24. Asrar J, Hill J. Biosynthetic processes for linear polymers. J Appl Polym Sci 2002;83(3):457-
483.
25. Ceccorulli G, Pizzoli M, Scandola M. Plasticization of bacterial poly(3-hydroxybutyrate).
Macromolecules 1992;25(12):3304-3306.
26. Riande E, Markovitz H, Plazek D, Raghupathi N. Viscoelastic behavior of polystyrene -
tricresyl phosphate solutions. Journal of Polymer Science Part C-Polymer Symposium
1975(50):405-430.
27. Ishikawa K, Kawaguchi Y, Doi Y. Plasticization of bacterial polyester by the addition of
acylglycerols and its enzymatic degradability. Kobunshi Ronbunshu 1991;48(4):221-226.
28. Withey R, Hay J. The effect of seeding on the crystallisation of poly(hydroxybutyrate), and
co-poly(hydroxybutyrate-co-valerate). Polymer 1999;40(18):5147-5152.
29. Alata H, Hexig B, Inoue Y. Effect of poly(vinyl alcohol) fine particles as a novel biodegradable
nucleating agent on the crystallization of poly(3-hydroxybutyrate). J Polym Sci Pt B-Polym Phys
2006;44(13):1813-1820.
102
30. Maiti P, Batt CA, Giannelis EP. New biodegradable polyhydroxybutyrate/layered silicate
nanocomposites. Biomacromolecules 2007;8(11):3393-3400.
31. Kosan B, Michels C, Meister F. Dissolution and forming of cellulose with ionic liquids.
Cellulose 2008;15(1):59-66.
32. Ljungberg N, Cavaille JY, Heux L. Nanocomposites of isotactic polypropylene reinforced with
rod-like cellulose whiskers. Polymer 2006;47(18):6285-6292.
33. Danyadi L, Janecska T, Szabo Z, Nagy G, Moczo J, Pukanszky B. Wood flour filled PP
composites: Compatibilization and adhesion. Composites Sci Technol 2007;67(13):2838-2846.
34. Nanocrystalline cellulose as strong as Kevlar and extracted from plants is heading
to commercialization in Canada. http://nextbigfuture.com 2011.
35. Lucia LA. The nanoscience and technology of renewable biomaterials. John Wiley and Sons,
2009.
36. Rasal RM, Janorkar AV, Hirt DE. Poly(lactic acid) modifications. Prog Polym Sci
2010;35(3):338-356.
37. Saeidlou S, Huneault MA, Li H, Park CB. Poly(lactic acid) crystallization. Prog Polym Sci
2012;37(12):1657-1677.
38. Nampoothiri KM, Nair NR, John RP. An overview of the recent developments in polylactide
(PLA) research. Bioresour Technol 2010;101(22):8493-8501.
39. Zaman HU, Song JC, Park L, Kang I, Park S, Kwak G, Park B, Yoon K. Poly(lactic acid) blends
with desired end-use properties by addition of thermoplastic polyester elastomer and MDI.
Polym Bull 2011;67(1):187-198.
103
40. Gui Z, Xu Y, Gao Y, Lu C, Cheng S. Novel polyethylene glycol-based polyester-toughened
polylactide. Mater Lett 2012;71:63-65.
41. Murariu M, Ferreira ADS, Pluta M, Bonnaud L, Alexandre M, Dubois P. Polylactide (PLA)-
CaSO4 composites toughened with low molecular weight and polymeric ester-like plasticizers
and related performances. Eur Polym J 2008;44(11):3842-3852.
42. Li H, Huneault MA. Effect of nucleation and plasticization on the crystallization of poly(lactic
acid). Polymer 2007;48(23):6855-6866.
43. Battegazzore D, Bocchini S, Frache A. Crystallization kinetics of poly(lactic acid)-talc
composites. Express Polym Lett 2011;5(10):849-858.
44. Yang G, Su J, Su R, Zhang Q, Fu Q, Na B. Toughening of poly(l-lactic acid) by annealing: The
effect of crystal morphologies and modifications. J Macromol Sci Part B-Phys 2012;51(1-3):184-
196.
45. Corre Y, Duchet J, Reignier J, Maazouz A. Melt strengthening of poly (lactic acid) through
reactive extrusion with epoxy-functionalized chains. Rheol Acta 2011;50(7-8):613-629.
46. Deenadayalan E, Lele AK, Balasubramanian M. Reactive extrusion of poly(l-lactic acid) with
glycidol. J Appl Polym Sci 2009;112(3):1391-1398.
47. Najafi N, Heuzey MC, Carreau PJ, Wood-Adams PM. Control of thermal degradation of
polylactide (PLA)-clay nanocomposites using chain extenders. Polym Degrad Stab
2012;97(4):554-565.
48. Villalobos M, Awojulu A, Greeley T, Deeter G. Oligomeric chain extenders for economic
reprocessing and recycling of condensation plastics, 2004.
104
49. Bikiaris DN, Karayannidis GP. Chain extension of polyesters PET and PBT with two new
diimidodiepoxides .2. Journal of Polymer Science Part A-Polymer Chemistry 1996;34(7):1337-
1342.
50. Villalobos M, Awojulu A, Greeley T, Turco G, Deeter G. Oligomeric chain extenders for
economic reprocessing and recycling of condensation plastics. Energy 2006;31(15):3227-3234.
51. Al-Itry R, Lamnawar K, Maazouz A. Improvement of thermal stability, rheological and
mechanical properties of PLA, PBAT and their blends by reactive extrusion with functionalized
epoxy. Polym Degrad Stab 2012;97(10):1898-1914.
52. Dean KM, Petinakis E, Meure S, Yu L, Chryss A. Melt strength and rheological properties of
biodegradable poly(lactic acid) modified via alkyl radical based reactive extrusion processes. J
Polym Environ 2012;20(3):741-747.
53. Rytlewski P, Zenkiewicz M, Malinowski R. Influence of dicumyl peroxide content on thermal
and mechanical properties of polylactide. International Polymer Processing 2011;26(5):580-586.
54. Carlson D, Dubois P, Nie L, Narayan R. Free radical branching of polylactide by reactive
extrusion. Polym Eng Sci 1998;38(2):311-321.
55. Zhang Y, Tiwary P, Parent JS, Kontopoulou M, Park CB. Crystallization and foaming of
coagent-modified polypropylene: Nucleation effects of cross-linked nanoparticles. Polymer
2013;54(18):4814-4819.
56. Parent JS, Sengupta SS, Kaufman M, Chaudhary BI. Coagent-induced transformations of
polypropylene microstructure: Evolution of bimodal architectures and cross-linked nano-
particles. Polymer 2008;49(18):3884-3891.
105
57. Parent JS, Bodsworth A, Sengupta SS, Kontopoulou M, Chaudhary BI, Poche D, Cousteaux S.
Structure-rheology relationships of long-chain branched polypropylene: Comparative analysis of
acrylic and allylic coagent chemistry. Polymer 2009;50(1):85-94.
58. Vaidya A, Pandey R, Mudliar S, Kumar M, Chakrabarti T, Devotta S. Production and recovery
of lactic acid for polylactide - an overview. Crit Rev Environ Sci Technol 2005;35(5):429-467.
59. Global biopolymers market analysis and forecasts to 2015. 2009;GMDCH0018MR.
60. Rai R, Yunos DM, Boccaccini AR, Knowles JC, Barker IA, Howdle SM, Tredwell GD, Keshavarz
T, Roy I. Poly-3-hydroxyoctanoate P(3HO), a medium chain length polyhydroxyalkanoate
homopolymer from Pseudomonas mendocina. Biomacromolecules 2011;12(6):2126-2136.
61. Nerkar M, Ramsay J, Ramsay B, Kontopoulou M. Morphology and mechanical properties of
blends of short-chain-length and medium-chain-length poly-3-hydroxyalkanoates. Polymer
Processing Society, 2012;358-359.
62. Akita S, Einaga Y, Miyaki Y, Fuita H. Solution Properties of poly(d-beta-hydroxybutyrate) .1.
biosynthesis and characterization. Macromolecules 1976;9(5):774-780.
63. Marchessault R, Okamura K, SU C. Physical properties of poly(beta-hydroxy butyrate) .2.
conformational aspects in solution. Macromolecules 1970;3(6):735-&.
64. Miyaki Y, Einaga Y, Hirosy T, Fujita H. Solution properties of poly(d-beta-hydroxybutyrate)
.2. light-scattering and viscosity in trifluoroethanol and behavior of highly expanded polymer
coils. Macromolecules 1977;10(6):1356-1364.
65. Cornibert J, Marchessault RH, Benoit H, Weill G. Physical properties of poly(beta-hydroxy
butyrate) .3. folding of helical segments in 2,2,2-trifluoroethanol. Macromolecules
1970;3(6):741.
106
66. Berger E, Ramsay BA, Ramsay JA, Chavarie C, Braunegg G. PHB Recovery by hypochlorite
digestion of non-PHB biomass. Biotechnol Tech 1989;3(4):227-232.
67. Ramsay J, Berger E, Voyer R, Chavarie C, Ramsay B. Extraction of poly-3-hydroxybutyrate
using chlorinated solvents. Biotechnol Tech 1994;8(8):589-594.
68. Shozui F, Matsumoto K, Sasaki T, Taguchi S. Engineering of polyhydroxyalkanoate synthase
by Ser477X/Gln481X saturation mutagenesis for efficient production of 3-hydroxybutyrate-
based copolyesters. Appl Microbiol Biotechnol 2009;84(6):1117-1124.
69. Daly P, Bruce D, Melik D, Harrison G. Thermal degradation kinetics of poly(3-
hydroxybutyrate-co-3-hydroxyhexanoate). J Appl Polym Sci 2005;98(1):66-74.
70. Jiang X, Sun Z, Marchessault R, Ramsay J, Ramsay B. Biosynthesis and properties of medium-
chain-length polyhydroxyalkanoate with enriched content of the dominant monomer.
Biomacromolecules 2012;13(9):2926-2932
71. Sun Z, Ramsay J, Guay M, Ramsay B. Enhanced yield of medium-chain-length
polyhydroxyalkanoates from nonanoic acid by co-feeding glucose in carbon-limited, fed-batch
culture. J Biotechnol 2009;143(4):262-267.
72. Jiang X, Ramsay JA, Ramsay BA. Acetone extraction of mcl-PHA from Pseudomonas putida
KT2440. J Microbiol Methods 2006;67(2):212-219.
73. Dupont A, Harrison G. Conformation and dn/dc determination of cellulose in N,N-
dimethylacetamide containing lithium chloride. Carbohydr Polym 2004;58(3):233-243.
74. Barham P, Keller A, Otun E, Holmes P. Crystallization and morphology of a bacterial
thermoplastic - poly-3-hydroxybutyrate. J Mater Sci 1984;19(9):2781-2794.
107
75. Beuermann S, Buback M, Davis T, Gilbert R, Hutchinson R, Kajiwara A, Klumperman B,
Russell G. Critically evaluated rate coefficients for free-radical polymerization, 3 - Propagation
rate coefficients for alkyl methacrylates. Macromol Chem Phys 2000;201(12):1355-1364.
76. Marchessault R, Monasterios C, Morin F, Sundararajan P. Chiral poly(beta-
hydroxyalkanoates) - an adaptable helix influenced by the alkane side-chain. Int J Biol
Macromol 1990;12(2):158-165.
77. Kim DY, Kim HW, Chung MG, Rhee YH. Biosynthesis, modification, and biodegradation of
bacterial medium-chain-length polyhydroxyalkanoates. J Microbiol 2007;45(2):87-97.
78. Ma P, Hristova-Bogaerds DG, Lemstra PJ, Zhang Y, Wang S. Toughening of PHBV/PBS and
PHB/PBS blends via in situ compatibilization using dicumyl peroxide as a free-radical grafting
initiator. Macromol Mater Eng 2012;297(5):402-410.
79. Ha CS, Cho WJ. Miscibility, properties, and biodegradability of microbial polyester
containing blends. Prog Polym Sci 2002;27(4):759-809.
80. Yu L, Dean K, Li L. Polymer blends and composites from renewable resources. Prog Polym
Sci 2006;31(6):576-602.
81. Zhang M, Thomas NL. Preparation and properties of polyhydroxybutyrate blended with
different types of starch. J Appl Polym Sci 2010;116(2):688-694.
82. Zhang M, Thomas NL. Blending polylactic acid with polyhydroxybutyrate: The effect on
thermal, mechanical, and biodegradation properties. Adv Polym Technol 2011;30(2):67-79.
83. Dufresne A, Vincendon M. Poly(3-hydroxybutyrate) and poly(3-hydroxyoctanoate) blends:
Morphology and mechanical behavior. Macromolecules 2000;33(8):2998-3008.
108
84. Martelli SM, Sabirova J, Fakhoury FM, Dyzma A, de Meyer B, Soetaert W. Obtention and
characterization of poly(3-hydroxybutyricacid-co-hydroxyvaleric acid)/mcl-PHA based blends.
Fod Sci Technol-Leb 2012;47(2):386-392.
85. Nerkar M, Ramsay J, Ramsay B, Kontopoulou M, Hutchinson R. Determination of Mark-
Houwink Parameters and absolute molecular weight of medium-chain-length poly(3-
hydroxyalkanoates). J Polym Environ 2013;21(1):24-29.
86. Sin MC, Gan SN, Annuar MSM, Tan IKP. Thermodegradation of medium-chain-length poly(3-
hydroxyalkanoates) produced by Pseudomonas putida from oleic acid. Polym Degrad Stab
2010;95(12):2334-2342.
87. Pearce R, Marchessault R. Melting and crystallization in bacterial poly(beta-
hydroxyvalerate), PHV, and Blends with Poly(beta-hydroxybutyrate-co-hydroxyvalerate).
Macromolecules 1994;27(14):3869-3874.
88. Owen A, Heinzel J, Skrbic Z, Divjakovic V. Crystallization and melting behavior of PHB and
PHB/HV copolymer. Polymer 1992;33(7):1563-1567.
89. Organ S, Barham P. On the equilibrium melting temperature of Polyhydroxybutyrate.
Polymer 1993;34(10):2169-2174.
90. Gagnon K, Lenz R, Farris R, Fuller R. Chemical modification of bacterial elastomers .1.
peroxide cross-linking. Polymer 1994;35(20):4358-4367.
91. Dufresne A, Reche L, Marchessault R, Lacroix M. Gamma-ray crosslinking of poly (3-
hydroxyoctanoate-co-undecenoate). Int J Biol Macromol 2001;29(2):73-82.
109
92. Park SJ, Kim TW, Kim MK, Lee SY, Lim S. Advanced bacterial polyhydroxyalkanoates:
Towards a versatile and sustainable platform for unnatural tailor-made polyesters. Biotechnol
Adv 2012;30(6):1196-1206.
93. Platel RH, Hodgson LM, Williams CK. Biocompatible initiators for lactide polymerization.
Polym Rev 2008;48(1):11-63.
94. Palade L, Lehermeier H, Dorgan J. Melt rheology of high L-content poly(lactic acid).
Macromolecules 2001;34(5):1384-1390.
95. Dorgan J, Williams J, Lewis D. Melt rheology of poly(lactic acid): entanglement and chain
architecture effects. J Rheol 1999;43(5):1141-1155.
96. Othman N, Acosta-Ramirez A, Mehrkhodavandi P, Dorgan JR, Hatzikiriakos SG. Solution and
melt viscoelastic properties of controlled microstructure poly(lactide). J Rheol 2011;55(5):987-
1005.
97. Othman N, Xu C, Mehrkhodavandi P, Hatzikiriakos SG. Thermorheological and mechanical
behavior of polylactide and its enantiomeric diblock copolymers and blends. Polymer
2012;53(12):2443-2452.
98. Othman N, Jazrawi B, Mehrkhodavandi P, Hatzikiriakos SG. Wall slip and melt fracture of
poly(lactides). Rheol Acta 2012;51(4):357-369.
99. Liu J, Lou L, Yu W, Liao R, Li R, Zhou C. Long chain branching polylactide: structures and
properties. Polymer 2010;51(22):5186-5197.
100. Pilla S, Kim SG, Auer GK, Gong S, Park CB. Microcellular extrusion-foaming of polylactide
with chain extender. Polym Eng Sci 2009;49(8):1653-1660.
110
101. Yu L, Toikka G, Dean K, Bateman S, Yuan Q, Filippou C, Tri Nguyen. Foaming behaviour and
cell structure of poly(lactic acid) after various modifications. Polym Int 2013;62(5):759-765.
102. Meng Q, Heuzey M, Carreau PJ. Effects of a multifunctional polymeric chain extender on
the properties of polylactide and polylactide/clay nanocomposites. Int Polym Proc
2012;27(5):505-516.
103. Najafi N, Heuzey MC, Carreau PJ. Crystallization behavior and morphology of polylactide
and PLA/clay nanocomposites in the presence of chain extenders. Polym Eng Sci
2013;53(5):1053-1064.
104. Maharana T, Mohanty B, Negi YS. Melt-solid polycondensation of lactic acid and its
biodegradability. Prog Polym Sci 2009;34(1):99-124.
105. Graebling D. Synthesis of branched polypropylene by a reactive extrusion process.
Macromolecules 2002;35(12):4602-4610.
106. Passaglia E, Coiai S, Augier S. Control of macromolecular architecture during the reactive
functionalization in the melt of olefin polymers. Prog Polym Sci 2009;34(9):911-947.
107. Yang S, Wu Z, Yang W, Yang M. Thermal and mechanical properties of chemical crosslinked
polylactide (PLA). Polym Test 2008;27(8):957-963.
108. You J, Lou L, Yu W, Zhou C. The preparation and crystallization of long chain branching
polylactide made by melt radicals reaction. J Appl Polym Sci 2013;129(4):1959-1970.
109. Fukada E. Poiseuille Medal Award Lecture: Piezoelectricity of biopolymers. Biorheology
1995;32(6):593-609.
110. Long Y, Shanks R, Stachurski Z. Kinetics of polymer crystallization. Prog Polym Sci
1995;20(4):651-701.
111
111. El Mabrouk K, Parent JS, Chaudhary BI, Cong R. Chemical modification of PP architecture:
strategies for introducing long-chain branching. Polymer 2009;50(23):5390-5397.
112. Wu W, Parent JS, Sengupta SS, Chaudhary BI. Preparation of crosslinked microspheres and
porous solids from hydrocarbon solutions: a new variation of precipitation polymerization
chemistry. J Polym Sci Pol Chem 2009;47(23):6561-6570.
113. Tian J, Yu W, Zhou C. Crystallization kinetics of linear and long-chain branched
polypropylene. J Macromol Sci Phys 2006;45(5):969-985.
114. Anderson KS, Schreck KM, Hillmyer MA. Toughening polylactide. Polym Rev 2008;48(1):85-
108.
115. Verhoogt H, Ramsay BA, Favis BD. Polymer blends containing poly(3-hydroxyalkanoate)s.
Polymer 1994;35(24):5155-5169.
116. Imre B, Pukanszky B. Compatibilization in bio-based and biodegradable polymer blends.
Eur Polym J 2013;49(6):1215-1233.
117. Schreck KM, Hillmyer MA. Block copolymers and melt blends of polylactide with Nodax
(TM) microbial polyesters: Preparation and mechanical properties. J Biotechnol
2007;132(3):287-295.
118. Noda I, Satkowski M, Dowrey A, Marcott C. Polymer alloys of Nodax copolymers and
poly(lactic acid). Macromol Biosci 2004;4(3):269-275.
119. Ma P, Spoelstra AB, Schmit P, Lemstra PJ. Toughening of poly (lactic acid) by poly (beta-
hydroxybutyrate-co-beta-hydroxyvalerate) with high beta-hydroxyvalerate content. Eur Polym J
2013;49(6):1523-1531.
112
120. Bhardwaj R, Mohanty AK. Modification of brittle polylactide by novel hyperbranched
polymer-based nanostructures. Biomacromolecules 2007;8(8):2476-2484.
121. Nyambo C, Misra M, Mohanty AK. Toughening of brittle poly(lactide) with hyperbranched
poly(ester-amide) and isocyanate-terminated prepolymer of polybutadiene. J Mater Sci
2012;47(13):5158-5168.
122. Semba T, Kitagawa K, Ishiaku U, Hamada H. The effect of crosslinking on the mechanical
properties of polylactic acid/polycaprolactone blends. J Appl Polym Sci 2006;101(3):1816-1825.
123. Wang R, Wang S, Zhang Y, Wan C, Ma P. Toughening modification of PLLA/PBS blends via
in-situ compatibilization. Polym Eng Sci 2009;49(1):26-33.
124. Ma P, Cai X, Zhang Y, Wang S, Dong W, Chen M, Lemstra PJ. In-situ compatibilization of
poly(lactic acid) and poly(butylene adipate-co-terephthalate) blends by using dicumyl peroxide
as a free-radical initiator. Polym Degrad Stab 2014;102:145-151.
125. Dong W, Ma P, Wang S, Chen M, Cai X, Zhang Y. Effect of partial crosslinking on
morphology and properties of the poly(beta-hydroxybutyrate)/poly(D,L-lactic acid) blends.
Polym Degrad Stab 2013;98(9):1549-1555.
126. Nerkar M, Ramsay JA, Ramsay BA, Kontopoulou M, Hutchinson RA. Determination of
Mark-Houwink Parameters and Absolute Molecular Weight of Medium-Chain-Length Poly(3-
Hydroxyalkanoates). J Polym Environ 2013;21(1):24-29.
127. Ma J, La LTB, Zaman I, Meng Q, Luong L, Ogilvie D, Kuan H. Fabrication, structure and
properties of epoxy/metal nanocomposites. Macromol Mater Eng 2011;296(5):465-474.
128. Abdousabets S, Patel R. Morphology of elastomeric alloys. Rubber Chem Technol
1991;64(5):769-779.
113
129. Palierne J. Linear Rheology of viscoelastic emulsions with Interfacial tension. Rheologica
Acta 1990;29(3):204-214.
130. Wu S. Formation of dispersed phase in incompatible polymer blends - interfacial and
rheological effects. Polym Eng Sci 1987;27(5):335-343.
114
Appendix A - Improved viscosity ratio and compatibility of poly (lactic
acid) and polyhydroxyoctanoate blends
Introduction
In Chapter 6 it was reported that the reason for the very coarse morphology of poly (lactic acid)
(PLA)/polyhydroxyoctanoate (PHO) blend is the significant viscosity mismatch between the
blend components. This is evident from Figure A.1, which summarizes the complex viscosities of
the materials under consideration. Both polymers had Newtonian behavior, with viscosities of
661 Pa.s and 13 Pa.s for PLA and PHO respectively, at 180°C, resulting in a viscosity ratio
(defined as the ratio of the viscosity of the dispersed phase over the viscosity of the matrix) of
0.02.
The shortcoming of viscosity mismatch between PHO and poly-3-hydroxybutyrate (PHB) was
addressed in chapter 4 by free radical mediated cross-linking using peroxide. In this appendix,
the same approach is followed to increase the viscosity of PHO and reduce the viscosity
mismatch with PLA. Additionally epoxidation of PHO was explored to improve its compatibility
with PLA by reaction of epoxy groups with hydroxyl and carboxyl groups of PLA. Furthermore it
was anticipated that the epoxy groups would act as a chain extender to improve the viscosity of
PHO [1].
115
Experimental
Cross-linking of PHO
PHO was crosslinked as described in Chapter 4 (Section 4.2.3) and dry mixed with PLA before
feeding in the DSM micro compounder. The gel content of the peroxide cross-linked MCL PHA
was measured as described in chapter 4.
Epoxidation of PHO
10-epoxyundecanoic acid was prepared as described below and was used to prepare epoxidized
PHO (ePHO). 50 g of 10-undecenoic acid were dissolved in 25 ml of anhydrous dichloromethane
and the solution was placed in an ice bath (0 oC). m-Chloroperbenzoic acid (mCPBA) was
purified by washing with a phosphate buffer solution of pH 7.5 and dried under reduced
pressure at room temperature. 65 g of purified mCPBA dissolved in 450 ml of anhydrous
dichloromethane were added to the 10-undecenoic solution drop wise, under continuous
stirring with a magnetic stirrer in an ice bath. The solution was then stirred for 24h until a
white precipitate formed. After filtration to eliminate the m-chlorobenzoic acid the filtrate was
washed with a 0.80 M solution of sodium sulphite until peroxide was no longer detectable with
peroxide test paper. The organic layer was finally washed with distilled water until pH test
paper indicated that the washings were neutral. The solution was then dried under reduced
pressure to yield a dry white powder.
Nuclear magnetic resonance characterization
Proton (1H) and carbon (13C) NMR spectra were recorded on a Bruker DPX300 NMR
spectrometer at 300 MHz in deuterated chloroform (CDCl3) as solvent. The chemical shifts
(ppm) for 1H and 13C NMR were referenced relative to tetramethylsilane (TMS, 0.00 ppm) as the
116
internal reference. The solution concentration was 0.5 w/v% (20 mg of sample in the NMR
tube) and the solution was filtered prior to placing it in the NMR tube.
Yield = 99%; mp = 39 oC. 1H-NMR (300 MHz, CDCl3, δ ppm); 1.9-1.3 (m, 12H, (CH2)6; 2.3 (t, 2H,
CH2-CO2H); 2.5 (dd, 1H, CH epoxy); 2.8 (t, 1H, Ha of CH2 epoxy); 2.9 (d, 1H, Hb of CH2 epoxy).
Results and Discussion
Cross-linked PHO (xPHO)
Complex viscosities of the xPHO, are summarized in Figure A.1. Cross-linking led to significant
increase in shear thinning behavior and loss of the Newtonian plateau as expected for cross-
linked polymer with high cross-linked densities.
Given the change in the viscosity-shear rate dependence, xPHO can only match the viscosity of
PLA in a very narrow frequency/shear rate range. As shown in Figure A.1, lower peroxide
loading (0.06 wt.%) matched the viscosity of PLA at lower frequency region. At higher peroxide
level, 0.2 wt.%, the viscosity of PHO was equivalent to PLA viscosity in the mid frequency
region. In order to match the viscosity of PHO at high shear rates of about 100 s-1 (equivalent to
shear rate in processing), 0.5 wt.% of peroxide was needed, hence PHO was cross-linked using
0.5 wt.% peroxide and blended with PLA for further evaluations.
117
Figure A.1 Complex viscosity of PLA, PHO, xPHO and ePHO at 180°C
The mechanical properties of PLA blends containing xPHO and ePHO are summarized in Table
A.1. xPHO matched the viscosity of PLA in the high frequency region and had pronounced shear
thinning characteristics, however, it did not show any improvement in mechanical properties as
the elastic nature of PHO was altered due to the high gel content (97%) of xPHO.
1 10 100
Co
mp
lex V
isco
sit
y (
Pa s
)
Frequency (rad/s)
PLA PHO 0.2 wt % Peroxide
0.06 wt % Peroxide 0.5 wt % Peroxide ePHO
100
101
102
103
104
105
118
Table A.1 Mechanical properties of PLA and blends containing PHO, ePHO and xPHO
Epoxidized PHO
The compositions of the prepared 10-epoxyundecanoic acid and the ePHO were determined by
1H NMR spectroscopy. The spectra were consistent with the expected structures. The
epoxidation reaction of 10-undecanoic acid was followed by 1H-NMR. During the epoxidation
the characteristic signals corresponding to the unsaturated side group (2.0 ppm –CH2-, 4.9 ppm
=CH2, 5.8 ppm –CH=) disappear and are replaced by peaks associated with the oxirane group at
2.5 ppm (c, 0-CH-, multiplet) and 2.75-2.9 ppm (a and b, -O-CH2, triplet and multiplet,
respectively), while the –CH2- group is now found at 1.3 ppm (e). The 1H NMR spectrum of
ePHA, together with the corresponding chemical shift assignments, is presented in Figure A.2.
The percentage of epoxy groups in the prepared ePHA was calculated by comparing the oxirane
signals (i and j in particular, due to the fact the h is overlapped by the CH2 side chain groups)
with the -CH3 signal of the unmodified side chain (f). The content of epoxy modified monomer
in the resulting ePHO was 12 mole %, as determined by the 1H NMR.
Tensile
stress (MPa)
Tensile
Strain (%)
Unnotched
Impact (KJ/m2)
Young's Modulus
(MPa)
PLA 74 (±3) 14 (±1) 32 (±5) 670 (±74)
10 % PHO 50 (±4) 35 (±15) 65 (±5) 696 (±29)
10 % xPHO 40 (±2) 11 (±1) 23 (±1) 567 (±29)
10 % ePHO 47 (±4) 32 (±8) 57 (±11) 551 (±36)
119
Figure A.2. NMR spectra of ePHA (12% mol/mol 10-epoxyundecanoic acid/unmodified
monomer)
Epoxidation of PHO did not show any increase in viscosity of PHO rather it was diminished
(Figure A.1).
Blends of PLA with ePHO
The compatibility of the blend components in immiscible blends is associated with their
interfacial tension. The Palierne model [2] was used to obtain an estimate of the interfacial
tension between PLA and PHO by fitting the complex modulus of the blend as a function of
frequency (A.3).
a
c
d
g
e
f
jh
b
c
d
b
i
a
b, h
f
d, e
c, g
i j
ji
d, h
a
Meth
an
ol
Ac
eto
ne
Chemical Shift ( )
120
The Palierne model relates viscoelasticity of emulsion with the viscoelasticity of matrix and
dispersed phase, droplet size and droplet size distribution of dispersed phase and the
interphase surface tension of the blend components. It can be expressed as
( ) ( ) ( ( )
( )) (A.1)
H(ω) =
(
)
(
)
Gd* and Gm* are complex moduli of the dispersed phase and matrix respectively, α, the
interfacial tension, φ, the volume fraction of disperse phase, ω is the studied frequency and R is
the particle radius.
Immiscible blends of PLA and PHO formed droplet-matrix morphology with PLA as matrix phase
and PHO forming spherical dispersed phase. Scanning electron microscopy (SEM) with details
mentioned in chapter 4 was used to characterize blend morphology. Volume average diameter
of PHO domain determined using image analysis software SigmaScan Pro was 1.73 mm.
Interfacial tension was calculated by substituting volume average diameter and volume fraction
of PHO in equation A.1. The estimated interfacial tensions were 2.1 and 0.6 N/mm for PLA-PHO
and PLA-ePHO respectively. The epoxy groups of ePHO plausibly reacted with carboxylic acid
end group of PLA improving their compatibility and hence reducing interfacial tension.
121
Figure A.3 Elastic modulus, G', of the PLA (matrix), PHO (droplet) and the 95/5 PLA/PHO blend
as a function of frequency at 180°C, and fit of the data using the Palierne model.
The mechanical properties of PLA/ePHO blend are depicted in Table A.1. Irrespective of
improved compatibility as a result of epoxidation, enhancement in properties was not
equivalent to that of neat PHO. ePHO that had viscosity lower than neat PHO, yielded a higher
viscosity mismatch and a coarser blend morphology (Figure A.4).
Figure A.4 Scanning electron microscopy of PLA blends containing 10 % of a) PHO, b) ePHO and
c) xPHO (0.5 wt.% of peroxide)
1.E-03
1.E-02
1.E-01
1.E+00
1.E+01
1.E+02
1.E+03
1.E+04
1.E+05
0.01 0.1 1 10 100
G' (P
a)
Frequency (rad/s)
Experimental G'
Palierne
Matrix G'
Droplet G'
20mm 20mm 20mm
a) b) c)
122
To further identify the mechanism for the morphology development the droplet size of PHO
and ePHO in PLA matrix was predicted using Wu model [3] for viscoelastic liquids as shown in
equation A.2 and further it was compared with particle size obtained from SEM images.
The model correlates viscosity of blend components, interfacial tension, droplet size and shear
rate according to
(
)
(A.2)
where, – shear rate, Ƞm- viscosity of matrix, Ƞd- viscosity of dispersed, σ- surface tension, D-
diameter of particle (+ve sign for λ>1, -ve for λ<1)
The predicted particle size at a shear rate of 100 S-1 was 4mm which is fairly close to the particle
size obtained from SEM images. The particle size of ePHO was higher at 8mm
In spite of the lower interfacial tension in ePHO/PLA, blend morphology was coarser, indicating
that effect of viscosity mismatch was more prominent than the interfacial tension reduction.
This explains the lack of improvements in morphology and hence mechanical properties.
Conclusion
Epoxidation of PHO improved its compatibility with PLA, however did not show any
improvement in elongation and impact properties due to increased viscosity mismatch with
PLA. Cross-linking of PHO improved its viscosity substantially. Irrespective of lowering viscosity
mismatch, xPHO/PLA blend did not enhance ductility, mainly due to high gel content altering its
elastic nature.
123
References
1. Al-Itry R, Lamnawar K, Maazouz A. Improvement of thermal stability, rheological and
mechanical properties of PLA, PBAT and their blends by reactive extrusion with functionalized
epoxy. Polym Degrad Stab 2012;97(10):1898-1914.
2. Palierne J. Linear rheology of viscoelastic emulsions with interfacial tension. Rheologica Acta
1990;29(3):204-214.
3. Wu S. Formation of dispersed phase in incompatible polymer blends - interfacial and
rheological effects. Polym Eng Sci 1987;27(5):335-343